Rotraninas > ‘ — vy Ww » | i l | = == % a 4 OLN ALL wre snp), seeHr ins a S ro x ay Ss? Hl \ ae oe SY PanOh? P 7. ee SE? ‘I iM ads Yn, a : he q } se io 1) Oe en oe aS ae nati, | ih O, Sy “4 | Lf, | Ty ee g ———e ——— OLS HIaS AL {> apatite te Me) is y NET ¢y Mats, Be i} fal } i tl Sy - ve tee fh 7 Vera i So a ir, eh amie ‘awa: en @ ty ng oS GR, 0 l HY \ YeES ! i O Nc t o “yy, a Ms Ay Sh Np Nos. 59, 60 ZC Ss October 31, 1957 S/ ATOLL RESEARCH BULLETIN 59. Report on the Gilbert Islands: Some Aspects of Human Ecology by René L. A. Catala 60. Climate and Meteorology of the Gilbert Islands by Marie-Hélene Sachet Issued by THE PACIFIC SCIENCE BOARD National Academy of Sciences—National Research Council : ® Washington, D. C., U.S.A. 29. ATOLL RESEARCH BULLETIN Se ot oe OS or an Ge on oe SO Ge ae oS om as om Report on the Gilbert Islands: Some aspects of human ecology by René L. A. Catala Climate and Meteorology of the Gilbert Islands by Marie-Héléne Sachet Issued by THE PACIFIC SCIENCE BOARD National Academy of Sétenéeatenetrod Research Council Washington, D. C. October (Bh aos 7: ACKNOWLEDGMENT It is a pleasure to commend the far-sighted policy of the Office of Naval Research, with its emphasis on basic research, as a result of which a grant has made possible the continuation of the Coral Atoll Program of the Pacific Science Board. It is of interest to note, historically, that much of the funda- mental information on atolls of the Pacific was gathered by the U. S. Navy's South Pacific Exploring Expedition, over one hundred years ago, under the command of Captain Charles Wilkes. The continuing nature of such scientific interest by the Navy is shown by the support for the Pacific Science Board's research programs, CIMA, SIM, and ICCP, during the past ten years. The Coral Atoll Program is a part of SIM. The preparation and issuance of Bulletin No. 60 was assisted by funds from the South Pacific Commission as well as from Contract Nfonr- 29154. “eo wo eae PelhleelUelUDe The sole responsibility of all statements made by authors of papers in the Atoll Research Bulletin rests with them, and do not necessarily represent the views of the Pacific Science Board or of the editors of the Bulletin. Editorial Staff F. R. Fosberg, editor M. H. Sachet, assistant editor Correspondence concerning the Atoll Research Bulletin should be addressed to the above c/o Pacific Science Board National Research Council 2101 Constitution Avenue, N. W. Washington 25, D. C., U. S. A. Reproduction in whole or in part is permitted for any purpose of the United States Government. ATOLL RESEARCH BULLETIN Report on the Gilbert Islands: Some aspects of human ecology by René L. A. Catala Issued by THE PACIFIC SCIENCE BOARD National Academy of Sciences——National Research Council Washington, D. C. Cctober Bley. PREFACE oh René L. A. Catala, the author of this report, ovtained his doctor's degree at the University of Paris. He is a graduate of the "Institut de France" and of the "Ecole Coloniale du Havre". The major part of his career was spent in Madagascar where he estab- lished coffee plantations and, later on, a private station for biological and morphological research. His work dealt in particular with the experi- mental variations of lepidoptera and the biology of river-fish. As a correspondent of the National Museum for Natural History in Paris, he carried out extensive scientific research in Madagascar. In 1944 Dr. Catala submitted to the Overseas Office for Scientific Research in Paris, a project for the establishiuent of a French Institute of Oceania in New Caledonia. Official negotiations, which he conducted himself in the United States of America, resulted in the assignment of buildings and equipment to that organization, on the basis of the Wash- ington Agreements of May 1946. At the newly created research centre, Dr. Catala, after having been in charge of the setting up of the I.F.0. services and the various research branches, undertook the study of coffee plantations in New Caledonia and coconut groves in the New Hebrides. After he was appointed head of the ecological laboratory, he specialized in the study of coral islands. Because of his special knowledge of ecology and his long experience of tropical crops, Dr. Catala was entrusted by the South Pacific Commis- sion with the survey which forms the subject of this report. The present report had its beginning in a research study carried out by Dr. Catala in 1951. The report of this study forwarded to the South Pacific Commission in French in 1952 was too extensive and detailed to permit publication. Technical experts of the Secretariat of the Comission have extracted the most essential contributions of the research and pvt them into suitable length for readers interested in Atolls. The final manuscript has been approved by Dr. Catala. During the intervening time since the original study was made, there have been numerous contributions to research oi Atolls, but the report which is presented here has much current value. The entire subject of Atolls is so extensive that the valuable original work of Dr. Catala will have much interest and worth to presenteday readers. Mr. Harold J. Coolidge, Executive Director of the Pacific Science Board, and a Member of the Research Council of the South Pacific Coimmission, has been of material assistance in obtaining publication of Dr. Catala's report in the well-known Atoll Research Bulletin, to which the South Pacific Comaission has contributed sowe financial support. Ralph Clairon Bedell Secretary-General South Pacific Commission Nouea, New Caledonia. Editor's Note: Because of discrepancies between the English inmanuscript as received and some of the original data, specimens, and the original French manu- script, as well as reidentifications of some of the material, it seemed desirable to edit the manuscript rather carefully. Miss Marie-Helene Sachet, assistant editor, has undertaken, with the author's permission, to do this, and has rather extensively retranslated and rewritten large parts of the manuscript. However, it was impossible to spend more time to give more uniformity to the style, and to reconcile throughout English versus American spellings and turns of phrase. No attempt was made to avoid using both English and metric systems of weights and ineasures. The aim was to make the report as intelligible and correct as possible. The arrangement, division into chapters, and reference to illustrations have been changed, also, fron the English version as submitted for publication. A few bibliographic references have been added.---Ed. nis PREFACE René L.A. Catala, the author of this report, obtained his doctor's degree at the University of Paris. He is a graduate of the “Institut de France" and of the "Ecole Coloniale du Havre". The major part of his career was spent in Madagascar where he estab- lished coffee plantations and, later on, a private station for biological and morphological research. His work dealt in particular with the experi- mental variations of lepidoptera and the biology of riverefish. Asa correspondent of the National Museum for Natural History in Paris, he carried out extensive scientific research in Madagascar. In 1944 Dr. Catala submitted to the Overseas Office for Scientific Research in Paris, a project for the establishiuent of a French Institute of Oceania in New Caledonia. Official negotiations, which he conducted himself in the United States of America, resulted in the assignment of buildings and equipment to that organization, on the basis of the Wash- ington Agreements of May 1946. At the newly created research centre, Dr. Catala, after having been in charge of the setting up of the I.F.0. services and the various research branches, undertook the study of coffee plantations in New Caledonia and coconut groves in the New Hebrides. After he was appointed head of the ecological laboratory, he specialized in the study of coral islands. Because of his special knowledge of ecology and his long experience of tropical crops, Dr. Catala was entrusted by the South Pacific Commis- sion with the survey which forms the subject of this report. The present report had its beginning in a research study carried out by Dr. Catala in 1951. The report of this study forwarded to the South Pacific Commission in French in 1952 was too extensive and detailed to permit publication. Technical experts of the Secretariat of the Comiission have extracted the most essential contributions of the research and put them into suitable length for readers interested in Atolls. The final manuscript has been approved by Dr. Catala. During the intervening time since the original study was made, there have been numerous contributions to research ou Atolls, but the report which is presented here has much current value. The entire subject of Atolls is so extensive that the valuable original work of Dr. Catala will have much interest and worth to present-day readers. Mr. Harold J. Coolidge, Executive Director oi the Pacific Science Board, and a Member of the Research Council of the South Pacific Coimnission, has been of material assistance in obtaining publication of Dr. Catala's report in the well-known Atoll Research Bulletin, to which the South Pacific Commission has contributed some financial SUBRORE- Ralph Clairon Bedell Secretary-General South Pacific Commission Nouuea, New Caledonia. Editor's Note: Because of discrepancies between the English manuscript as received and some of the original data, specimens, and the original French manu- script, as well as reidentifications of some of the material, it seemed desirable to edit the manuscript rather carefully. Miss Marie-Helene Sachet, assistant editor, has undertaken, with the author's permission, to do this, and has rather extensively retranslated and rewritten large parts of the manuscript. However, it was impossible to spend more time to give more uniformity to the style, and to reconcile throughout English versus American spellings and turns of phrase. No attempt was made to avoid using both English and metric systems of weights and measures. The aim was to make the report as intelligible and correct as possible. The arrangement, division into chapters, and reference to illustrations have been changed, also, from the English version as submitted for publication. A few bibliographic references have been added.-~--Ed. aD TABLE OF CONTENTS Page INTRODUCTION | . , IV CHAFTER 1 PHYSICAL AND CLIMATIC CHARACTERISTICS 1 2 THE SOILS h 3 THE PEOPLE Lh 4 THE COCONUT PALM 21 PART I: THE COCONUT PALM IN THE GILBERT ISLANDS 22 II: THE COCONUT PALM AND THE CLIMATE 30 III: THE COCONUT PALM AND GROUND-WATER 35 IV: CULTIVATION OF THE COCONUT PALM 36 V: USES OF THE COCONUT’ PALM yh 5 THE PANDANUS ita) 6 THE BREADFRUIT TREE 61 7 THE BABAI 67 8 OTHER PLANTS USED BY GILBERTESE 76 9 VEGETATION AND FLORA 79 10 INSECTS COLLECTED IN THE GILBERTS noes 11 MARINE RESOURCES 118 PART I: GENERAL SURVEY 118 II: DEEP-SEA FISHING 118 TII: REEF AND LAGOON FISHING 122 IV: COOKING METHODS 134 V: TRADE U's) VI: FISH PONDS 136 VII: MISCELLANEOUS MARINE RESOURCES 139 12 DOMESTIC ANIMALS he 13 QUALITATIVE STUDY OF GILBERTESE DIET yah 14 GILBERTESE HANDICRAFTS A5e 15 CO-OPERATIVE SOCIETIES 15h. 16 CONCLUSIONS AND RECOMMENDATIONS 158 PART I: A PROPOSED PRCGRAMME OF AGRICULTURAL IMPROVEMENT 158 II: EXPLOITATION OF MARINE RESOURCES 165 III: MISCELLANEOUS RECOMMENDATIONS 167 GILBERTESE VOCABULARY 171 LIST OF REFERENCES 163 APPENDIX I 187 Tilt INTRODUCTION This report concludes the survey assigned to my wife and me by the South Pacific Commission in connection with Project E.6, under the general direction of Dr. H. G. MacMillan, Executive Officer for Economic Develop-~ ment. The survey covered a period of six months of field work in the islands of the Gilbert Group, from 6 March to 30 August 1951. The approved objectives were, "firstly, to study and assess the physi- cal environment, particularly the soil, water, and climatological relations; the vegetation in its economic or other relationship; the production of the sea; in a word, the 'ecology' of this representative area. A second ob- jective was the improvement of the human ecology of the area, by discovering ways and means of increasing the quantity and variety of subsistence and comuercial crops, and of achieving greater unit production by improved fer- tility and method; the improvement of domestic animals; and the improvement of fisheries and other products of the sea." We were favoured in that we arrived in the Gilbert Islands after two years of severe droughts and just before the beginning of a period of very abundant rains. We were thus able during the first stage of our visit to study the effects of drought and, a few months later, those of almost daily rain. We were fortunate enough to be able to visit almost all the islands in the Group except Kuria, Maiana, Nonouti, Tabiteuea and Tamana. But these gaps, however regrettable, were compensated by the knowledge we gained of neighbouring islands, with identical or very similar general conditions. We inay consider ourselves lucky to have had an opportunity of visiting eleven islands within a few months, in view of the often considerable distances between the various island groups and of the very simall number of ships ply- ing between them. Our visit to some of these islands was very short, but nearly always long enough to form a good comparative picture in relation to those on which we stayed longer; Abemama, Nikunau and, of course, Tarawa, where we established our headquarters, and lived in two localities, at Bairiki and Bikenibeu respectively. Although we made the most of this six months' visit and recorded as many observations as we were able to, a most interesting, if not essential, section does not appear in this report, namely, the ecology of coral reefs and lagoon waters, of which it was impossible to make a real study for lack of adequate equipment, boat and fishing gear. However, there were valuable compensations for these deficiencies, and it gives us great pleasure to mention here the unceasing assistance extended by the administrative authorities of the Gilbert Islands, under the dis- tinguished direction of the Acting Resident Comuissinner, Mr. Ph. Keegan. The Resident and his staff took a continued interest in our survey, its IV objectives and implementation, and thanks to their unfailing kindness, we found upon our arrival many imore facilities and much better accommodation than we could have expected. We were able to visit a number of territories, sometimes being invited by the Resident to accompany him on his official visits, sometimes taking advantage of very fortunate "coincidences" in shipping movements - not always due to chance. Records likely to be of in- terest in connection with our survey were made available to us in the spirit of cooperation displayed by everyone who could, from personal experience, supply us with information. Finally, we were able to hold frequent meetings which always helped in elucidating problems or advancing the implementation of the project. . We must also mention the valuable information supplied by the Tarawa health authorities, by Captain Brown, mv. "Tungaru", by Captain Reiher, landowner in Abemama, and by several missionaries who either had spent their whole lives in the Group and were thus well informed on a number of subjects, or who had long studied every aspect of life in these low islands. Father Sabatier, Father Zufrey and others were mines of information. Great benefit was also derived from our conversations with school teachers and other native civil servants and from the cooperation of a few natives; some of an older generation who still have much useful knowledge on their country, and others who have adhered to their traditions through keeping in close contact with their environment. Our colleague, Mr. R. Mason, an officer of the Fiji Department of Ag- riculture, for his part concentrated his activities on a limited area of Tarawa Atoll. Considering the short time he spent in the Group, his obser- vations were recorded in a very comprehensive, well-illustrated report. Neither must we omit to pay tribute to our friend, Monsieur J. Barrau who paid us a short visit. By his knowledge of soils, his ability as an observer, and his critical sense, he made a most valuable voluntary contri- bution to our survey. Acknowledgement should also be made of the very valuable assistance given by various individuals and institutions in identifying specimens. The following specimens were sent to the Australian Museum, Sydney, for identification: Fish to Mr. Gilbert P. Whitley, crustaceans to Mr. F. McNeill, molluscs to Miss Joyce Allan, other marine specimens to Miss Eliz- abeth Pope, a few insects to Mr. A. Musgrave. The plants were identified in a preliminary way by Professor A. Guil- laumin of the National Museum, Paris, and by Dr. Marcel Bauman of the Botanical Museum, University of Zurich. Sea-weeds were sent to Mrs. Valerie May (Jones) of the Sydney Botanical Gardens. Algae from fish ponds (Cyanophyceae) were sent to Dr. Drouvet of the Chicago Natural History Museum.* Macromycetes and micromycetes, mosses, lichens and land algae were sent to Professor R. Heim of the National Museum, Paris. Lepidoptera were sent to Monsieur Pierre Viette of the National Museum, Paris. A consign- ment of insects belonging to various groups was sent to the Bishop Museum “See Appendix Teper. in Honolulu and preliminary lists of identifications were received from the following specialists: Miss Amy Suehiro, Mr. R. H. van Zwaluwenburg, Dr. D. Elmo Hardy, Dr. 0. H. Swezey, Mr. E. H. Bryan, Jr. Finally, some plank- ton specimens were sent to Professor Monod, Head of the "Institut Francais d'Afrique Noire", Dakar. Unforeseen contingencies and long delays in shipping which were beyond our control involved a considerable delay in the delivery to various spe- cialists of several cases of specimens; the corresponding identifications will be accordingly delayed. An addendum to the present report will prob- ably be necessary. Except for specimens known to us beyond any possibility of doubt, this report will often show, after the names indicated, the genus . or family only. All local names were checked several times and, apart from a few men- tioned as doubtful, even in native circles, they may be considered accurate. It should be noted that the collections made in a variety of fields by American experts of the Pacific Science Board working in Onotoa*, cannot fail, in due time, to add valuable complementary data to our more general inventory. It will be readily understood that, for our part, we concentrated mainly on specimens of economic value, having a direct bearing on the sur- vey assigned to us. In conclusion may the author also be allowed to pay tribute to the invaluable assistance given by his wife, who constantly helped in all spheres of this survey. Rene L. A. Catala Noumea, New Caledonia June, 1952. STEERS *See Atoll Research Bulletins nos. 12, 13, 28, 47 and 57. vi Q 10/20 30 40 50 60 70 '!80 90 100 Ram Nonouti Fig. 1 - The Gilbert Islands CHAPTER 1 PHYSICAL AND CLIMATIC CHARACTERISTICS Formerly known as the Kingsmill Group, the Gilbert Group straddles the Equator and extends over 400 miles from the island of Arorae in the south to the island of Little Makin in the north-west (Fig. 1). It constitutes the central link in a long chain of islands stretching from the Marshall Group in the north to the Ellice Group in the south. f , f The Group lies between 3°30 latitude north and 2°45 latitude south and between 172°30 and 177° longitudes east. It may be divided into eleven atolls and five reef islands. (The atoll is a chain of small islands set around a lagoon, while the reef island has no lagoon and is washed on all sides by the ocean. However, both were formed by coral growths on a pedestal of submarine peaks. These islands are very low, rarely exceeding 3 meters above the aver- age tide level.) Figures 2 and 3 show some of the essential differences be- tween atolls and reef islands. The Gilbert Islands may be divided into three groups: North - Little Makin - Butaritari Centre - Marakei - Abaiang - Tarawa - Maiana - Abemama - Kuria - Aranuka South = Nonouti - Tabiteuea - Beru - Nikunau - Onotoa - Tamana - Arorae These sixteen atolls and islands represent a land surface of 114 square miles, distributed as follows (Pusinelli, 1947): Name of Island Sq. Mile Population density per sq. mile } ity ath Ni 1947 Little Makin 2.80 345.0 Butaritari 4.50 4O5.1 Marakei 3.94 437.4 Abaians, 11.06 255.4 Tarawa Toes W535 Maiana 10.39 AGT eS Abemama Cn SH) 173.7 Kuria 4.98 63.3 Aranuka 5.971 Gta Nonouti 9.682 203.9 Tabiteuea 19.00 199.2 Beru O25 273.8 Nikunau 7.00 227.4 Onotoa ip ae BBO.) Tamana 2.00 WIL Arorae 5.00 Bu 750 Total Tae nee Average BENS ee an Bie i it Ie) oe ye = Josie Sak pak reskin el ermine iinet sc arr cama Sana ee aS iva Am OAS ee ea al 3 s Tan i 9 Te é i 14 ’ : ae t - ‘ : i { mec a aA nln onan ape inten aoe 4 Sst = i The most recent population figures supplied by the Administration of the Gilbert Islands at 3lst December, 1950 are presented in Table II. The high demographic density of these populations clearly demonstrates that the Gilbertese is pérfectly adapted to the conditions obtaining in this environment, where he makes use of almost all the resources offered, at least from the variety, if not from the quantity, point of view. One may add that these living conditions, daily problems of food supply, and animal and plant resources, have obviously created in this race of such varied origins, a characteristic Gilbertese pattern of customs and physical and mental life. From the physical point of view the Gilbertese are generally robust, with excellent muscular strength, but with little resistance to prolonged strain. In spite of the drought which had just occurred when we arrived, most of them were rather fat - a result of the high nutritive value of toddy, according to local medical officers. The same medical officers state that health in the Gilbert Islands is generally good, that tuberculosis is the most serious disease and that the comparatively high tuberculinic index shows the absolute necessity of methodical detection and would justify not only the carrying out of a con- tinuous programme, but also a great increase in the number of medical of- ficers. It is typical that investigations carried out in 1929 already showed, in the Gilbert Islands, an estimated figure of 30% of deaths from tuberculosis alone. The head of the Health Department in Tarawa also gave a very high figure of positive reactions. However it is not our task to discuss this question here. The Gilbertese are clean and healthy, the more so from being clad sensibly, that is to say in keeping with the climatic conditions of their country. They are extremely sensitive to the slightest fall in temperature, which explains the fragility of their respiratory system. It seemed to us that natives wearing only the ancestral riri, and whose bodies are continually exposed to sun and wind, show less sensitivity than those who are over- clothed or who - mostly women - because of acquired prudishness, go as far as swimming or fishing with all their clothes on, afterward remaining for a long time exposed to the inevitable cooling influence of the almost constant breeze. From the psychological point of view we neither wish nor pretend to analyse the Gilbertese mentality after staying only six months with them. A stay of many years and a thorough knowledge of the language would be re- quired for that. We shall, however, give some brief personal impressions which may -throw light on the comments we shall have to make in this survey on the behaviour of the Gilbertese when confronted with certain problems of an economic nature, The Gilbert Islanders generally appear to show sharpness only when de- fending their ancestral rights to land plots, which is quite logical. But they are not so astute regarding money, not enough, one might even say. See They like money, of course, but spend it with remarkable facility, and too often on items of doubtful usefulness. However, they do not like begging (with the exception of the bubuti custom practiced among natives) as do so many primitive peoples in other parts of the world. On the contrary, when they give a present because custom requires it, they are really vexed if a return gift follows so soon that it might be regarded as a payment. For the pleasure of being hospitable or of demonstrating sincere feeling, they do not hesitate to spend sums of money which are large in comparison with their small salary or budget. We are embarrassed when we think of the cost of the telegrams of good wishes spontaneously sent to us in New Caledonia by Gilbert Islands natives. The Gilbertese are proud, but very approachable, and pleasant to deal with. Their native distinction often finds expression in kind actions which are all the more touching from their being nearly always pexfectly disin- terested, Another aspect of their nature is their total confidence in others, both in moral and material dealings. We also appreciated their independent spirit and their frankness, which is often disarming. Their answers, whether posi- tive or negative, are always direct. But the Gilbertese' forthrightness does not preclude a form of respect devoid of obsequiousness. His often un- expected reactions are never arrogant, and are a corollary of his independent, individualistic nature, as are his teasing spirit and fanciful mind. Both are expressed in choreographic attitudes, in which mimicry always has a deserved success. Finally, these people have a highly-developed artistic sense, and it would be difficult to find anything to equal some of their extraordinarily beautiful choral singing. It is really in their dances and choral singing only that the Gilbertese express the whole genius of their race, and can give rein to an exuberance which, because of a surprising modesty, is no longer manifested in the ordinary course of their everyday life. The Gilbertese are an intelligent people. Many show real pride in hav- ing risen above the general level, but it did not seem to us that this was ever expressed in a contemptuous or even haughty way. Those working with Europeans are generally avid to learn and to understand everything, and are full of gratitude for whoever may have increased their knowledge, even about their own territory. The bubuti custom will at times be mentioned in this report. It plays too important a part in the individual economy of the Gilbertese to be passed over entirely. It consists of asking a man of means, or a man who suddenly finds himself well supplied with money, various items recently ac- guired or even with gifts, for some part of his riches. We were told that there are three degrees in bubuti. At the first, the request can easily be evaded. This is more difficult at the second, while at the third, one has to meet the demand. One among many other examples: a Gilbertese comes home after one or two years' work with the British Phosphate Commissioners. He has earned & pounds per month, with practically no expenses. He brings back a tidy sum of money, various items bought recently such as cloth, a sewing alae SS machine, a bicycle, fishing gear, watches, fountain pens, and stick tobacco. Close and distant relatives eagerly welcome him and, through the bubuti, pro- ceed to take what they fancy. The opulent one is in a short time stripped of practically everything. One of the victims told us that the only way of avoiding this family octopus was to state that such an item had been bought for the father, another for the mother, and so on; in this way, the owner can still share with them the use of the "reserved" items. It is thus easy to understand how money is bound to be squandered, and how amounts, which if kept intact, would have a high purchasing power, do not retain any great value when shared among so many spongers and can be used only for futile purchases at the neighbouring store. With regard to the physical activity of the Gilbert Islander, assuredly he is not lazy. There is no laziness in a people who in the past carried out such hard work as the digging and excavating of babai pits with extremely sketchy implements, and who today still spend most of their time actively seeking their food supplies at sea, from the soil and at the top of coconut palms. Neither are the Gilbertese indolent in the sense of being sluggish or apathetic. They are indolent simply in the sense that they adjust them- selves to the climate in which they live. Like all peoples living in sandy countries, they rest during the hot hours of the day, and stretched on the floor of their open-sided huts, wave to the white people passing through the village, thinking them eccentrics for cycling for miles when the sun is so high. They ae to spend nights fishing on the reef flats by torch or lamp light. ( Peta onal ck al ester | bey (1) The Gilbertese is secure in his home environment, there are usually adequate coconuts. and pandanus on his lands aud fish in the sea to meet his frugal dietetic requirements and by cutting copra to provide for his needs from the store. There is little need or incentive to accumulate wealth for its own sake. Their mental approach to life is "why work if you don't have to?"; they are perfectly content to sleep the days away. On the other hand, and as Dr. Catala points out, they are equally prepared to work really hard, as has been shown by the excellent reputation they have earned working phos-. phate and stevedoring ships at Ocean Island and in the Labour Corps during the war. Well led - not driven - they can be most industrious. It is, therefore, certain that as workers away from their homes they are as good as most, but that at home their needs are too simple and their way of life too. traditional to be able to induce them to work industriously. A further factor which should be mentioned since it retards commercial development, is the reluctance of a Gilbertese to work for or to teach another. This is a result of excessive independence and individualism and means that a large landowner is unable to employ others to work his lands for him and is only able to get members of his family to work them. To a certain extent the master/servant problem is being overcome by employment through the Co-operative Societies, but the reluctance to teach each other is an even more restrictive practice and means that an expert cultivator, or fisherman or maker of handi- crafts will not spread his knowledge and is only prepared to pass it on to his family shortly before his death. This: problem will only be overcome by more advanced education. (Note by the Administration of the Gilbert and Ellice Islands Colony. ) ye The Gilbertese Community It seems necessary to add some brief information on the Gilbertese com- munity in order to describe in greater detail the background of the various activities covered in this survey, and because the Gilbertese necessarily have the prevailing if not the exclusive role in such activities. Villages Villages (fig. 13) are spaced out at distances which vary considerably according to the islands. The houses stand on both sides of the road, or on one side only when there is too little space between the road and the lagoon. These villages comprise groups of from fifteen to fifty huts, sometimes more. The dwellings, built entirely of waterial from the coconut palm and pandanus, are most often slightly raised on coral blocks brought from the reef flats on the ocean side. In the southern islands, this elevation is occasionally very great (Plate IIa). Some of these huts have an additional story (Plate IIb), which is reached by a small ladder. Generally, each house shelters one family only. In an open space in the village, and standing out among the other build- ings because of its size, the maneaba, or community house, is the meeting place of the notables - the "Old Men" - and it is used more generally for all the group manifestations of native life. It also serves as a temporary shelter for visitors. For others it is merely a road-post. These buildings sometimes are of considerable size, as shown in the photograph of the Little Makin maneaba (Plate IIc), one of the largest native buildings in the Gilbert Islands. In building these maneabas - as indeed any other house, or a canoe, even of the largest size - not one nail or screw is used. Mortise and tenon are never used and pegs are the exception. All main and secondary parts are assembled and held together with coir cord (Plate IId). Other buildings in the village include canoe sheds and canoe-building shelters. Finally, a few pigsties are located in the shade of trees, outside the village on the inland side. Some are built of coconut trunks, others of landing strip matting, a reminder of the war (see Plate XIITIa). On the lagoon side, buildings are set on piles. Some, simall and of fragile construction, support a floor used for fishing at high tide, when they are completely isolated in the water. Others, more strongly built, jut out over the water, and are connected with the shore by means of a frail gangway of coconut logs. These are Gilbertese privies, but they are some- times used for several ends. They are something of a salon, incomers and outgoers exchanging small talk with the occupier, while a fishing enthusiast who has been unable to find room on the fishing platforms is quite happy to use this excellent promontory (at high tide, of course). The village is kept extremely clean and enlivened by flowers planted around the houses. They are mostly imported plants, some valued for their fragrance, others for the beauty of their blooms and still others because elie eae a ee ie = SS they do not wilt easily. This last point is important since the making of leis and ornaments for dances holds a very large place in native activities. In the villages of the southern islands, and in particular on Nikunau one sees, usually around the huts, side by side with pumpkins, numerous tobacco plants. The dried leaves provide a rather poor substitute for stick tobacco when vhe latter is lacking or there is no money to buy it. Finally, plants growing to a good height also shade or decorate the Gilbertese village: coconut palms, pandanus, breadfruit trees, pawpaws, tamanu, and a few trees or shrubs of no economic value as poincianas, acaly- phas and oleanders. Land Teaure Almost all the land in the Gilbert Islands belongs to natives and cannot be sold to anybody else. The establishment of land registry has made real progress only since 1939, when it was decided to appoint a Land Commissioner, while in 1948 a native land court was set up to assist the Lands Commission in clearing up an extremely confused situation caused by incessant disputes among natives, the confusion being made still greater by the extreme fragmentation of properties. However, this land registration work is sajd,to be virtually completed in seven of the sixteen islands in the Group. 1) We were told that in the past the Gilbertese did not live in villages, but that each family lived on its land (or rather on one of its pieces of land). It is possible that the inhabited plots were much better tended than they are now. We know that the authorities, far from opposing a return to past conditions, would possibly be ready to encourage the natives to live on their properties, but it is unlikely that, being now used to community life, the natives would, of their own accord, return to the earlier scattered mode of living. Emigration A young civil servant in Betio told us half seriously and half in jest: "The best produce of the Gilbert Islands is the Gilbertese, considered as labour outside the territory". With the exception of Ocean Island and Nauru, which do not take him far from his environment and where the general condi- tions are very similar to those in his own territory, we do not think it would be such a good thing for the Gilbertese to migrate too far and to countries where general conditions are too different from his own. (1) There is no central Land Register, and each island keeps its own Regis- ter. As regards the Lands Commission, it would be more correct to say that real progress was first made between 1935 and 1937, when Mr. Kennedy completed settlements on four Ellice Islands and Mr. Maude carried out a settlement on Ocean Island. However, Dr. Catala is dealing only with the Gilberts and I ao not think it can be claimed that real progress was made in that Group un- til the Commission was reconstituted in 1947. (Note by the Administration of the Gilbert and Ellice Islands Colony. ) anoe We were told that many of them were always ready to go, not out of necessity but curiosity, and because upon their return they would have an obvious superiority over the others. They like to travel - just like many people everywhere - rather for the avtraction of things new than from any compulsion. Therefore, the word "exodus" would not correspond, in reality, to any intention or imperative need. However, greater numbers of Gilbertese might leave if some among them were specialized in a trade, but few or none are. The most recent absentee figures made available to us were: at the phosphate works, 452 men; at Christmas and Fanning Islands, 150; and a dozen scattered on Canton and other islands. Nearly all of them are contract labourers who are only temporarily away from their home islands. The number of women and children accompanying them totals approximately 350. -20- CHAPTER 4 THE COCONUT PALM GENERAL The coconut palm, Cocos nucifera L. reaches 20 m. or more in height. Like all monocots, it has no tap root, but a radiating, much developed root system, the roots being constantly renewed. This is the reason wh; the palm.is so flexibly attached to the ground, and so resilient to winds. Authors are not agreed on the number of leaves of a normal tree; 25 to 35 seems a reasonable figure. It is generally accepted that it is necessary for the coconut to grow in a light, deep soil, indispensable for the good development of the root system. Standing water is nefarious and so is impervious substratum. The coco- nut may be regarded as a halophilous plant, because of its tolerance of salinity, but salt is not indispensable. There is a limit to this tolerance, beyond which the survival of the plant is endangered. The needs of the coco- nut palm for atmospheric humidity and rainfall are as great as for ground- water. It can be grown successfully on all sorts of soils, including the very highly calcareous ones of the Pacific coral.islands. Flowers: The inflorescence is enveloped in a sheath, the spathe. Toward the end of its growth, this splits along its whole length and the spadix, covered with buds, appears. The larger buds are female fiowers, the smaller, male. The flowers of both sexes mature at different times, which prevents auto- fecundation. Fertilization is realized by crosses between trees. Fruits: A longitudinal section of a mature coconut shows from outside in: A - epidermis ‘D - kernel including oily endosperm and B - mesocarp (fibrous husk) small cylindrical embryo C - Endocarp (shell) E - cavity which contained the Liquia(1) Productivity: It is usually accepted, that, under optimum conditions, the first fruit appears about the 6th or 7th year. Actually, under normal con- ditions, a planted coconut tree begins useful production about the 10th year, and full production about the 15th year. In the best physico-chemical conditions, and under cultivation aid care, a coconut may remain productive until its Oth year and even further. Life-span: Generally, the productive life of a dente palm can be compared to human life in length. (1) This liquid, rich in potash calts, is erroneously referred to as “coco- nut milk"; it would be much more accurate to call it "coconut water", be- cause true "coconut milk" is a white milky liquid expressed from grated fresh coconut kernel and used as an indispensable ingredient in the prepar- ation of many Pacific island foods and sweetmeats. (Ed.) wows Yield: Various authors, especially J. Adam (1942) give as generally accepted as normal, averages of 5,000 to 6,000 nuts for one ton of copra, with extreme figures of 3,500 and 8,000. oon also indicates as normal a copra production of 5 to 8 kg per tree per year. 1) It must be kept in mind that these figures refer to planted coconut trees. PART I: THE COCONUT PALM IN THE GILBERT ISLANDS Except for the very small number of coconut trees truly planted in the agricultural sense of the word, the immense majority of the palms covering the Gilberts often form a real forest rather than a plantation. This does not invalidate the remark of Turbott (1949), that most of the coconut trees of the Gilberts were planted by man. If planted means putting in place a sprouted coconut, then it can be accepted that the coconut forest has been at least in the beginning planted by man. This intervention has been less obvious as times passed and now there is a much greater proportion of spon- taneous trees than "planted" trees. This will be better understood further on when the "cultivation methods" for coconuts are studied. The height of old coconut palms ranges from sixty to eighty feet. The latter figure is rearely exceeded. Generally, the palms along the ocean shore, and even more those along the lagoon shore tend to lean outward, while those in the interior are straight. Some authors think that this leaning is not due to the wind, but to a tropism to light. The intense reflection of light from the water, especially on the lagoon side, and from the sand, ine tensify this effect. The palms on the ocean side show a much less abundant leaf system than those in the centre and on the lagoon side. There are seldom more than twenty fronds. The fronds are much damaged by the spray which "burns" part of the pinnae, especially during drought periods, and the palms have a fairly consistently lower production of unuts. The coconut palms in the central area form thick stands practically everywhere. These stands are not always only of tall palms. Young coconut palms, occasionally in large numbers, grown from nuts sprouted where they fell, are found with palms of all ages and conditions, as are also representae tives of species of trees, shrubs and low plants. This jungle generally starts forty to fifty metersback from the rampart (ocean side) and covers an area proportionate to the width of the island. It is sometimes so dense that it is difficult to walk through (Plate IIIa) especially when it is also choked up with heaps of dry fronds and the trunks of dead palms. On the lagoon side the "forest" of palms becomes slightly thinner. This is mostly due to the road common to all these islands, and to the clean- ing which is done on either side of it to a distance of 8 to 10 m. The strip (1) A coconut palm, under optimum conditions, is capable of producing well over 100 nuts a year, but, at 55 palms per acre, an annual yield of 4,500 nuts per acre may be taken as a very good plantation average. (Ed. ) ape jove of land between the road and the lagoon is sometimes very narrow, only a few metres wide. The coconut palms sheltering it are generally less close to- gether than anywhere else. Qur first impression on arrival in these islands was extremely painful. Except in a very few privileged areas, the coconut palms reflected tragical- ly the effects of an intense drought which had lasted almost two years. Over considerable areas on many islands, even on those of the northern group reputed to be the least affected, the rusty~yellow leaves seemed burned by the sun. Others, dry and still attache d, were hanging dowi the trunks which were sometimes entirely covered (Plate IIIb). The almost complete absence of nuts added to this sad picture. Most palms either did not bear any nuts at all or they had extremely small or atrophied ones. Nevertheless, an un- opened spathe appeared here and there. Only palms near the villages which they shelter, and some others under very exceptional conditions, still bore just enough coconuts to deserve their names. The return of the rains enabled us, less than two months later, to form a much better impression. Almost daily rain, often very abundant, rapidly changed the dismal picture into a comforting sight; the fronds were green again; inflorescences were numerous but generally did not set fruit. SP Seine. inflorescences which then followed in rapid succession, also failed to set, as could be seen from the abundance of small nuts, te nimoimoi, slightly bigger than hens' eggs, strewing the ground in greater numbers every day. Later inflorescences in spite of continuing heavy rains were more success- ful, and less than six months after the beginning of the rains, most trees bore numerous nuts at various stazes of development. Thus, our last impression of the Gilbert Islands was the opposite of the first. Without appearing over-optimistic, we could give a most favour- . able forecast for the coconut crop to be gathered in under a year, even in _ the islands of the Southern group. Although there, the recovery was notice- ably late, and perhaps proceeded more slowly, the rains were abundant and distributed over a long period; they were not completely over when we left and it was generally felt that the westerlies, expected to blow in October, would again bring some more rain, making a good crop more certain. One of the characteristics which force themselves on the attention of an observer when examining coconut palms in the Gilbert Islands is the extra- ordinary development of the root system high above ground level (Plate IIIc) found in most trees. This phenomenon illustrates the need for these trees i)" iets interesting to note that the same observation was made after the great drought of 1924-1925. In the "Handbook for Atoll Research", Pacific Science Board, National Research Council, May 18th, 1951, p.10, appear these few lines taken from the Annual Report for these years (no. 1369, 1 28): "The drought then broke and the coconuts which had ceased to bear, put out fresh spathes of blossom. When the spathes opened, however, rains fell with such torrential force that the young blossom was beaten from the stalks and, though the palms throve, the first coconut crop was a failure." aie to build a sort of pedestal of roots to strengthen their base because the nuts they come from simply rooted where they fell or were not planted deep enough. Such development practically never occurs on palms grown from seed- lings transplanted under good conditions. Root formation along the trunk occurs also in cases of natural layering of coconut palms which for various reasons have bent towards the earth, and then started growing upright again (Plate IIId). Such cases are not exceptional in the Gilberts and even such grotesque trees have great vitality and bear many leaves and nuts. A great many trees have at their bases and slightly above ground a aun- ber of additional roots, issuing from breaks in the bark. These breaks may be due to the influence of a parasite, but it is also possible that they may be the result oi the natural growth of the roots, intended to strengthen the palms. This consolidating might be rendered necessary by ‘erosion of sandy places where winds can easily blow the soil away. Such hypotheses are con- cerned only with a small number of cases. Still other trees show at heights up to 2m. ringlike swellings, with young roots appearing where the bark cracked. Other palms had lost their bark all around the stipe on a width of several dm., and were encased in sleeves of matted roots. We have io satisfying explanation for these phenomena, which may be due tc parasitic effects. Similar root production is the rule where the bark has been damaged by fire. We observed a very large number of palms that should have been felled, as they were too old, sterile or so rarely productive that their existence was not justified. These useless palms either occupy space necessary for younger specimens, or crowd productive palms. Often the density is so high that the palms get in each other's way both at the base and summit. Some are choked, others cannot spread, and this condition can only lower the gen- eral yield, especially when the tangle is increased by the luxuriant and usually Ge jsestabiibu growth of the shrub and tree species of the under- growth. - The question of spacing seems to us to be of great importance. In some areas we were able to verify to what extent anormal spacing, resulting in a reasonable number of coconut palms for a given area, could be beneficial to the trees. Good examples are the palms of the Makin School in Abemama and those on the site where the future Bikenibeu school is to be built. In spite of the irregularity of their distribution (Fig. 14), the latter will be given as an example of normal density. Table V lists the number of fronds and nuts borne at a given time by each of these 133 palms. The total number of palms in the whole area is more significant than the excessive proximity of some of the palms aw the base; in fact it often happeus that when the bases of palms are almost touching their summits are quite far apart. (1) New Gilbert Islands Regulations will provide for the establishment of Agricultural Committees. Through these organisations it is intended to plan a drive on cutting out over-age trees. (Note by the Administration of the Gilbert and Ellice Islands Colony.) 12) Nes | very eiroydoztyy pue sTydmeg fe) | | | J 2 eee eee noose] aa eee 1 ' 2 Pemphis and Rhizophora Area CL” ppm= i280 jl 1 1 t Bikenibe Eita 3 o 9 Ic ie 87 i) teas : mas i rr 120 uo 115 ° 8 *o ue 7 18 2 oO 2 { 4 +5 ° | : IP 130 o: n3 5 ie 0 i CLE ppm| =s80 ° Sig 32131 tS) 0° all : ° ° ° 109 x x ° ‘ 2 as jos 107 3 ° ° 2 2 103 104 ine pbm= 130 a fo2z | ClUippm= 460 r | x ae ! ; | ; | | : : a - solu | ° 2 (98 bs S j a0 94 if I ; 92 . | S 91 | GI 90 ° q o 86 87 86] f é 120M | iom ee oe ESS is ina -O : CL ppm =48 CL ppm =32 e2 83 65 i 5 ° i I 2 8 ! le CL ppm = ice Wes 2: 79 rs iP 72 7s. te ° ; i 30/M i “ps 2 oo T Ie 70 ° Ss ; | see | = 62 a09. os 64 __| 6s. —— ! 60 | x él + * Cloppm = +de/200 ° é, 57 r ap o 59 58 ° x " CLoppm = 288 + ° ‘ sr cry ss = I 54 x x e283 a7 46 ° 50 an és ¥E 49 ° lo 40 a é 39 y ae 33 32 Lo] esos Ts Ec) x 27) 28 : 7, 2 $ ° i. ; : : 25 4 ao 1 To 138 Bearing Coconut Palms : 2 = eo 0 Young Palms 1 . This needed adjustment of the crowns is realised readily thanks to the easy bending, curving and later straightening up of the palms. But this balance can be achieved only if the palms are approximately of the same age and if they are not too numerous, which unfortunately is not always the case. The most frequent picture is that of coconut palms of all ages; the youngest suffer from too much shade and cannot develop unless old palms die and make room for them. But only a few palms profit by this natural replace- ment; while a great many more are hampered by excessive density. Ata dis- tance from such tangles, areas will often be found where the coconut palms have been destroyed by fire and the plants of the bush have spread, forming shrubby vegetation. These observations apply to a lesser or greater extent to all islands in the Group. Coconut palm varieties The natives of the Gilbert Islands recognize some varieties showing really different characteristics. In Tarawa for instance, where we were able to make a careful study of these characteristics, the natives first divide coconuts into two categories according to whether the mesocarp is edible (te bunia) or inedible (te ni)(1), a characteristic of some palms and not of certain stages of the development of the fruit (Plates IVa and IVb). In each of these groups the coconuts are then arranged according to the colour of the epidermis, two colours only being taken into consideration: green and orange. The varieties with edible mesocarp are: Te bunia uraura - pale yellow-green with orange tinge Te bunia roro ~- pale green The varieties with non-edible mesocarp are: Te ni uraura ~ yellow-green orange Te ni roro - true green A few essential characteristics of the varieties will be found in Table III. Unfortunately we did not have enough nuts at our disposal and had to be content with three nuts at the same stage of ripeness (moimoto), which were however gathered on palms typical of each variety and from the same locality. It would indeed have been difficult to ask the natives to take too many nuts from their individual stock, which was limited after the drought. (I) Synonym Te ni: the coconut palm in general. “254 A few remarks concerning these varieties are necessary: fo. 1. It seems possible to connect the "te bunia” with their edible meso- carp with che "Saccharina" variety of Miquel (cf. Baas Becking 1949). 2. The use of colour only as a criterion makes it rather difficult to separate varieties especially in the case of "te bunia uraura" and "te ni uraura", which have almost the same colour. A specialist would have to study in greater detail this delicate question of coconut varieties, and we recommend that such an expert should visit the Gilbert Islands. Gilbertese terms for stages of coconut development The importance of the coconut palm in the life of the Gilbertese people is such that they have a comparatively rich vocabulary for it. For each Stage of growth of the palm or the zut there is a special term, some of which terms will be used in the course of this survey. It should be added that some confusion seems to prevail in the works of some authors on the exact meanings of certain words in this special vocabulary. It can be noted (Fig. 15) that the arrangement of the various growth stages of the nuts corresponds somewhat to a normal growth curve. We re- corded the weight of nuts and quantity of liquid found at each successive stage. It may be noted that the greatest volume of liquid corresponds to the moimoto stage, and that the nuts are gathered precisely at this stage if they are intended for drinking purposes. The same table finally indicates the pH measured by colorimetry for the water of four nuts at each stage. The palm itself would be called on Tarawa, for instance, by the follow- ing names: Te: raka - Germinated nut Te uto - Stemless coconut seedling with ouly a tuft of fronds visible above ground. Te ene - The palin when the trunk is beginning to ap- pear. Te kaikai - A palm which has not yet borne flowers or fruit. Te ai - Productive coconut palm. Te ni nikawai - Very old palm bearing only a few small fronds. Abnormalities Three abnormalities may be mentioned: I. Inflorescences are sometimes found which are not ramified and form a simple spike instead of several as is the rule. This anomaly, called ne mori, (Plate IVc) involves all the inflorescences of a single palm. The nuts reach maturity. This is quite certainly an abnormality, not a variety. 2) Ses qnu JO UIDTM qnu jo qy2ue7 i ee gee ee a | See oN s 2, Y alt 3, By a) Hy Da ae a ees Se eee as ee ee ee eee) ee ee i> Cpessoecoss aouarzeayUMIITO Trays jo yqysueT uaets) aBuezo MOTTAR |~ user) aBuerIoO MOTTA Feet ured Bunok JO styoer Jo 1OTOO AIOAT 4InIZ Bunok KIA useTB VUSTT MOTTOEX ueeis qyuSstT Aro, ZSsouTe MOTTOA 7USTT pue sdUsDSaIOTJUT JO IOTOD 65h GOS oft Eze T6E OE HEE Oge 92h GLt Ort Tu ut sayem jo AyTVueNnv 260°2 COiine —0cl-¢ €9S°T oS6°T Gon *e olG’e ~SLE*e G9c‘e to9fe Sore But 4 fnrZ STOyM JO yUBTOM 0°OT 0°OT 0°6 G°g 8°8 0°6 8°s 0°6 G°OT §=S°OT S°OT wo UT Treys JO 4IPTM e°Tt Star Q°TTt G*OT Q°Tt O°TT S°OT S°OT Gore SPO S°OT wo UT TIays JO yysueT O£* 0% 0€" 09 O€° oF Gd Ole Goes OlmEGe OF (So Cae (e159 Cab (o\S9 eh} (Og? CHL Cre OB —COr72(O) (ed G2°0O G2°0 G2°0 G2°0 S2°0 Gc°O SE°0 = GE" GE*O wo Uf; dxB.0pus JO sseuyoTUL Ee 0°?2 JEECA OS ore H°€ NPE OAS AAS wo Ut dresosewu jo ssouyoTyL O°ET 0°ST O°2eT G°ST 6°9T €°ST OE S°Ge O°ST Wo UT qna JO UIPTM 8°qT SGI Q°qT G°ST €°cT 0°ST SME GPCHe G°ST wo UT qynu jo 7OsTSH L°Qt Q°LT O°T2 9°Te O° ed h° od G6t = 9 6E O’°eT wo UT qnu fo qysueT G° Lt 0°0S G° hh G°0S Game O° St ONG GORAS oq Wo UT Y4nuU JO soUaTAaTUMDITO 0°0f 0°O€ 0°0€ G°6¢ o°Te o'Té OO cu OLOS 0°0€ WO UT € asptz jo yysueT 1°92 0°0£ €*62 G*6¢ G*0€ G°0€ 62 0208 0°0f uo UT g@ espTI JO YVsusT €°62 €°62 €°6é2 G*°62 0°0€ G°of Geicme Gele 1°92 wo ut T esptt jo yysueT AZSTIBA YoRo € 2 fi € Z T ¢ z T yo sanu € JO suoTsueuTg SsBuBIO UTM asuero anu 3y3 JO 9seq 343 48 aBue1O UTM pesutazy ‘uaary pesutazy ATpeoriq ‘user | AOTTEA UTA pesutay ‘user UTM pesutay users yreq| soepTede, queqstszed jo 10TOD aseq out 72 eseq oui. 78 uses aseq oud 28] S14 TUM ysouTe ‘a8uT7 eSuero STULIapTds JO IOTOD uaety aad ‘esuero MoTTef-use.zr5 a4TUAM ysouTe ‘usea3 4USTT ULTIMA UdssIB=MOTTOL VST T Oxy ce eee ee ee serieTeh TU 33% :dies0sem STqTpeut YLTM SaInn etung 04 :d2ed0SemM STqtTpe YFTM SIN SoT 1070789 “TTIOVW Bere], UO synuos0D JO seTYeTIeA “TIT WIaVL — ee ——- whet 2 an Emad y == eet ee ee oar iw preg ci Patel e eM, ¥ ie rare = hots ai. seiaale x 2 . “ + ree he ee hele oft as ae Cary “a ¥ *sqnu ay JO W.AOIS Jo saBeys QuUeTEZJTP ay} JO soTYstTaayoereyO “TIO eMere], ,CTOY TN OL, Adetrea wred ynuos0g GT °St¥ *soeTd JuetessFp Be ut SupMoz upped Sree pee eae qUetleyJTp B MOTT sured YnU YOSY °oeteqs yoo 4q3u ZoJ synu AMoZ Buren ‘1z0yBoOTpuT TeszeaTun pewgiee ee 6 se per TAQZel UTM pemmsvom seam Ad suz, SITZSUT}US STqnd UT 13eqZBA JO A4TQUEND O2\c SOS Spe sp Ss pe OFS ES 000€ 0002 Ja1eM 4nuoD0D jo Ad surety UT 4UZTeM °T saMoup uUEeD UT suoTsueuTq es - oh aap = ee i ete es : Saltire. Od ped STAM ce ‘i aatiy II. Te wae (Plate IVd) are aspermous nuts, without hard endocarp, in which the fibrous mesocarp occupies all the space with the exception of an extremely reduced central cavity. These nuts are always very long. A cur- ious fact is that the saiie palm may bear at the sai:e time bunches of te wae and, from other inflorescences, bunches of norial and fertile nuts. Iii. There is a teratological case which, while extremely rare, should nevertheless be mentioned, first because of its strange character, and second because of its connection with the harbourage of pests. The first characteristic is that the leaflets cannot open because on one side there is a strip which fits into a slot on the opposite side of the leaflet. The second characteristic is that each leaflet shows in its proximal part two successive folds at a very acute angle. The aspect of this fasciation (Plate Va) reminds one strongly of the Elaeis with "welded leaf- lets" (Blaeis Dybowskyi Hua), called "Fadé" in Dahciey, but this is only a comparison of general appearance since we have here a sterile teratological coconut palm and not a variety fixed and fertile as for "Fadé". The term "welded leaflets" is inappropriate in both cases, because the leaflets are normally formed secondarily by the splitting of the blade, and here remain wore or less united. Such palms should be ruthlessly destroyed since they are homes for pests, each closed leaflet sheltering a great quantity of Decadarchis and other iticrolepidoptera not yet identified. Dwarf Coconut Palm - Coco Niiio A few trial plantings of dwarf coconut paliis have been wade from sisall mudbers of nuts: three dozen in 1949, six dozen in 1950, and quite recently again about the same number. It is, of course, impossible to draw any con- clusion whatever from the few young palms grown from the two first consign- ments. These had been planted too recently, and were very widely distributed, under very varied conditions. Pests Generally speaking it can be stated that coco..ut palus in the Gilberts are at present free from the worst pests, except for those listed below and also a very tiny psychid which causes very little damage, and an undeter- mined borer which burrows in the blade of the leaflets and pupates under a fine silk layer near the iiain vein. Although these larvae can be very numerous on come trees, they do not seem to affect them. The same is true of the trees carrying a rather large number of Coccidae. The palms in the Gilberts are free of Brontispa and rhinoceros beetle. Lepidoptera: Decadarchis sp. (Lyonetiidae). Common in all the coconut groves of the Gilberts, but the damage does not seem cause for worry. These insects are Bone never very numerous on one tree, except in the aberrant palms whose fronds do not unfold, where they occur in extraordinary abundance, due to the pro- tection they are afforded. The palms immediately around these may be more infected than those far away. Coleoptera: Diocalandra frumenti (F.) (Rhynchophorinae). This weevil, although fairly generally distributed in the Gilberts, does not create the kind of spectacular damage wrought by the closely related species D. tahitensis (Guérin-Meneville) in other regions of the Pacific. It was never found in great numbers, but in groups of a few individuals under the bark of the trunk, where the larvae and pupae are easy to find. They were often found on small nuts fallen on the ground who had failed to set (te nimaimoi). Pentodon or Papuana sp. (1 \scaranaeddneh This beetle has been tentatively identified as Papuane. hubneri Fairm. by M. R. Mason (of Suva), as Pentodon australis Blackburn by Bryan and by Swezey as a sitall relative of the co- conut borer (Oryctes sp.?). More specimens are evidently needed, we had only one male and one female, sent to Honolulu. The natives say that this insect bores into the tubers of babai and it was found easily by us (and later by R. Mason in the same part of Tarawa) on the bases of banana plants. The insect is said also to attack young coconut paliis but we were unable to verify this. P. Lepesme (1947) says that Papuana voodlarkiana Montr. (for - which he gives many sysonyus, but no P. hiibneri) is reputed to attack young coconut plants, burrowing into the ground and destroying the plants at the base of the stem. Orthoptera: Graeffea cocophaga (New.) (formerly called Lopaphus), Phasmidae. Native name: te rokati rokati, pronounced rokass is undoubtedly a distortion of "locust"). This insect has caused much damage to palms on Abemama and is a aed threat for the rest of the area. Damage: The leaflets of the fronds are in the beginning marked by indenta- tions at first rather far apart. As the blade is eaten between them, there remains finally only the vein. The eggs laid by insects clinging to the hanging part of the frond roll in the trough formed by the main vein of the frond and those of the leaflets, and fall on the ground at varying distances from the tree. Some imst roll in the opposite direction, that is, toward the base of the frond, and make control of the pest difficult. Many weeks after the laying, the eggs hatch and the young insects climb the nearest palm trunk. (I) According to Dr. 0. L. Cartwright Pentodon australis Blackb. is now regarded as a synonym of Metanastes vulgivatus Olliff, found in Lord Howe Island, New South Wales and South Australia. Papuana hiibneri Fairm. is found on Duke of York Island and New Britain. They are not especially restricted in food habits. - Ed. Sous Extension of Graeffea cocophaga New. on Abemama Atoll. YOM, | ‘ Areas already attacked in (after Sabatier) apace History: In Sabatier, 1939, p. 36 we read: "Two pests are prospering: a long-bodied locust imported from Hawaii, which eats the leaves of the coconvt palms in Abemama...." (the second pest is the beetle discussed above). We met the author, of the Sacred Heart Mission, now living on Tarawa after 40 years on Abemaima, who gave us some valuable additional information: he told us that Graeffea is said to have been introduced from Hawaii or from Kusaie to be shown to the natives as a curiosity about 10 years before his arrival, therefore about the beginning of this century. He wrote: "I - the pest which eats the leaves of the coconut trees had its center in the last village to the north, Tapiang, and stretched for 3 to 4 kilometers along the road. II - Another point attacked was a little to the north of the village of Takatirika, the second village from the north. III - A third point at the south end, in an islet called Kena. There may be other points of lesser importance." Present conditions: During our stay in Abemama we tried to make as complete a survey as possible. Here are the results: I - Graeffea infests the atoll from the northern tip near Tabiang to a few hundred meters beyord the village of Tabontebike. In the village itself at Tabontebike, it is interesting to note that the palms were not attacked by Graeffea, while less than 20 meters away on the other side of the road a great many trees had many visible parasites. Perhaps the smoke from cook- ing fires and some other factors disturb the insects in the village. II - At the other end of the atoll, the center of infestation observed by Sabatier on Kena islet has extended to the next islet and spread for about 2km. Kena itself is completely infested. It seems that all over this islet, the damage is worse than anywhere else (in percentage of trees and fronds damaged). The map (fig. 16) speaks for itself. To justify the importance we give to this pest we can quote other authors. Froggatt (1914) quotes Wyatt Gill on an invasion of this pest on the Hervey Islands "An invasion of these voracious insects is almost as much dreaded by the Islanders as a plague of locusts in the East. - I have seen immense groves of coconut palms destroyed in a few months by this species of phasma...". Simmonds and Copeland also report cases of sudden propagation of this insect, which we had seen ourselves in other regions of the Pacific (Catala, 1948). In Abemama the infestation is not as much of a catastrophe as that described by Wyatt Gill, but the fronds of the palis are damaged and the yield of nuts must be affected. It can always be feared that a combination of favorable circumstances may someday provoke a more serious invasion by this pest. We do not know any enemies of it, but there may be a parasite controlling it. It can especially be feared that Graeffea nay migrate to other parts of the archipelago thus far all free of it. No one can predict the possible results of such a spread. When we landed on Ocean Island, we observed on a large whale boat taking labour and freight ashore from the Tarawa ship, an acridian which, in the absence of accurate identification, would appear to be of the genus Catantops. During the next few days, we were to observe the very serious damage caused to the coconut palms of the island by considerable numbers of uaoe the same insect. This "locust" has not yet been mentioned in the territory of the Gilbert Islands proper, and we have not observed it there. Miscellaneous: The coconut crab (Birgus latro) does not cause very extensive damage. On the contrary, the rats do, and the damage is in proportion with the density of the plant jungle and also with the density of the palins of the variety te bunia with edible husks. In places we saw sone tin sleeves in- vended to keep the rats from climbing the trunks, but this device is often not very efficient as the fronds of the palm to be protected touch those of other trees and rats can get from one tree to another this way. The natives do not seem to try any more active control, and it seems that a minimum of care would substantially reduce the number of rodents. As for cryptogamic infections, they did not appear serious except for a limited area on Abemaina where many leaflets were infected by Pestolozzia. PART II: THE COCONUT PALM AND THE CLIMATE Most of the manuals on the coconut discuss the optimum set of conditions for a rational and profitable cultivation. None of it applies in the Gilbert Islands. The coconut palms of low coral islands and those of the Gilbert Islands in particular, grow under special climatic and edaphic conditions, which, it should be said, are often mediocre, and they cannot behave like those giving a very high and more regular yield in more favoured regions of the Pacific where comaercial cultivation prevails. On the other hand, the considerable size of the coconut groves and the great hardiness of the co- conut palm in the Gilbert Islands make up for the poor conditions. Response to Drought: While as regards rainfall one has to divide the archi- pelago into three groups, north, centre, and south, it would seem that as re- gerds the behaviour of the coconut palm, these distinctions are less sharp, and might even be reduced to one group for the northern and central islands and another for the southern islands. It has also been indicated that the islands of the southern group were affected by droughts more than those in the centre and north. However, these latter sometimes pay a very heavy price for droughts, although the human population of these territories is not too severely affected. From the extensive information we were able to gather on the spot, it appears that the production capacity of the coconut palm may be retained on an average for eighteen months after the beginning of the drought period. In other words, not only does the lack of rain have little bearing on the future of inflorescences which have already formed, but spadices continue to appear, and the ripening of the nuts will not necessarily be affected. Their size may decrease considerably (see Table IV of weights for Nikunau) but the meat will be well-developed and the quality of the resulting copra will not be affected. This “prolonged productivity" cannot be predicted and depends in fact upon conditions varying considerably from one place to another under the influence of basic factors (soils, moisture, degree of salinity of the ground water, etc.). It also varies from one palm to another according to individual characteristics. =u TABLE IV Size and weight of 50 Coconuts grown during period of drought on Nikunau, and the weight in ounces of the freshly cut coconut meats. Length in Circumference Whole Nut Meat, fresh Nos. inches in inches Wit au OZ. Wt. in Oz. a 9 ss 8 3 2 9 sphalsy 9 h 3 8 10.5 6 2 y 8.5 11 10 h De Ge Ss) 5 2 6 8 10 735 3 4 8.5 11.5 10 25) 8 7.5 11 7 3 9 9 iy 11 2.5 10 8.5 12 Lh. 2.5 ae 7.5 10 7 3 12 6.5 9.5 45 Aloe 13 8 8 B55 2 1h I O25 t 3 15 9 10 Ff 15 16 ¥ (is) 3) hos 5 17 8 B25 6.5 3.5 18 8 10 9 5 19 Te 9 6 2 20 8 11.5 10 2 21 7 10 if 3.5 22 8.5 1a 1a 3.5 23 7 10 ii 5 eh 8 9 5 2.5 25 8 12 1a h 26 i 10.5 Te 355 27 8.5 all 9 h 28 8 12 13 KAS 29 9 13 il 3 30 8 10 ©) B05) 31 if ae) if 3 ge i S35 6 3 33 8 1255 le 2 34 8 at 6 3 35 7 10 6 3 36 8.5 go! 10 3 37 8 1025 3 3 38 8 10 it 3 39 T 9.5 6 3.5 LO 8 10 8 5 hy 8.5 10 - O 3 TABLE IV (continued) Circwaference Whole Nut Meat, fresh Length in Nos inches in inches EA, Wt. anOze 0) Wh,” dm Oz.- 42 7.5 10 7.5 3 43 aes es 5 asada BIST hb 8 12 10 h 45 8 10.5 Ou Sa 46 Gea 8 6.5 2 L7 8.5 9.5 7 3 48 8.5 9 6 2 ho 8.5 ie) 9.5 2 50 8.5 9 Oui: 2 Total Weight fresh meats = 143 oz. or 4.054 Ke. . Total weight dry meats Reonen) - 91 oz. or 2.580 Ke. Loss of weight on drying - 36.35 per cent 19,330 coconuts required per ton of copra at Nikunau, Southern Gilberts, during period of drought, 1951. As drought continues, the fresh water reserves in the sub-soil decrease and in the most serious cases they may eventually be exhausted. Such factors as sunlight and wind become adverse because they increase the loss of water from the leaves. The palms become less productive and finally not at all. But such is the extraordinary resistance of the palm, that even after weather- ing more than two years' intense drought and in the most affected areas, it is still able to produce inflorescences. The natives use these for the essential production of karewe (toddy) and stand in no danger of curtailing copra pro- duction by doing this, since, in most cases, the flowers would not have set or would have produced abortive nuts without any commercial value whatsoever. It is obvious that the effects of drought are not felt everywhere in a territory with the same acuteness, and that the palms react differently ac- cording to age, individual characteristics, and, most of all, according to their location. Thus on each island there may be seen to a very great degree a replica of what was observed on another: coconut palms either killed by drought, affected to a high degree, or untouched or barely affected. A study of environmental conditions corresponding to these three categories of be- haviour, and setting forth the relations from cause to effect, provides valuable information. I. PALMS DYING FROM DROUGHTS They are found under the following conditions: (a) Palms on extremeties of islands, and on narrower strips of land between ocean and lagoon are particularly exposed to wind and salt spray. The ground level of corners of islands is, in some places, noticeably lower than the ridges parallel to the ocean and lagoon sides. The layer of sand is generally very thin, and the roots of the coconut are unable to penetrate very deeply, owing to the proximity of the rocky platform, and are especially close to sea water. The same conditions are realized when the strip of land around the lagoon is especially narrow (as between Bikenibeu and Bonriki on Tarawa). The promontories located at the edge of channels through which water enters the lagoon with each incoming tide and flows out at the ebb, are frequently unstable either from erosion by friction, or from deposition of considerable amounts of sterile coral shingle. All these factors contrive to render precarious (Plate Vb) the existence of the most marginal coconut palms and it is easy to understand that, in time of pro- longed drought, they cannot resist. The same applies to other vegetation. The loss of these trees is relatively unimportant as they are of low productivity; they are mostly useful as wind- breaks .(1 (1) In the northern Marshall Islands, under similar conditions, the Marshal- lese commonly leave a strip of native vegetation to serve as a windbreak or shelter the plantation from wind and spray, and to prevent erosion. - Ed. etl (bo) (a) (e) Gravel zones, which are found at many points on most islands, com- prise a surface layer of coarse coral shingle of varying thickness. This accumulation of detrital elements absorbs a considerable amount of heat, resulting in intense evaporation; this is all the more noticeable as, for lack of a cover of low vegetation, the immediate underelayer drains too rapidly. In normal times, the coconut palm grows with difficulty in these areas. It is quite obvious that it cannot resist when the drought extends over long periods. The occurrence of a hard substratum always constitutes a very im- portant factor in the behaviour of the palms especially if this layer is so hard that roots cannot penetrate it. When it is ata relatively high level but still sufficiently deep for the root system to develop freely, the coconut palm draws upon the fresh water reserve accumulated above the hard layer. In times of pro- longed drought this reserve is soon exhausted and the palm cannot survive. In addition to these usual causes of destruction of the coconut palm, accidental factors may also occur which have serious con- sequences only because they coincide with a period of extensive drought. Thus, considerable areas may be seen around the Marakei Lagoon where all the palms have died (Plate Vc) probably through the following combination of circumstances: Although on a level only slightly higher than the highest water- mark of the lagoon, these palms had successfully resisted previous great droughts as witnessed by their size and their age (not less than thirty years). They had of course been severely affected and each time must have been very near the limit of tolerance to salinity. The same thing happened again during the last drought, and then came an exceptional tide reaching a hitherto unrecorded level. Some’ even maintain that it was a tidal wave. In any case, the whole of the area was flooded for a few hours. Fresh-water reserves were practically non-existent and, as there was no com- pensatory correction for this sudden and massive excess of chlorides, the palms died in a large numbers. Destruction of coconut. palms by fire is even more common, and affects much larger areas in time of droughts, increasing as the droughts become more intense. The hanging dry fronds of the palms are a great fire hazard (see below and Plate IIIb). Too often we saw large areas devastated in this way (Kena Island, Abemama, etc.). PALMS AFFECTED BY DROUGHTS The great majority of palms on the Gilbert Islands are affected, more or less, by the effects of prolonged droughts. We know that under normal con- ditions, the transpiration rate of the leaves is considerably higher among the fully grown ones, that is to say in the age group between the eighth eee month after their expansion until the first signs of natural wilting. Therefore, it is the oldest leaves which use up most of the water reserves available to the palm. When a long period of drought arrives these re- serves diminish, and the coconut palm gets rid of the now superfluous leaves at an accelerated rate. It is perhaps the very rapidity of this desiccation which makes these fronds hang from the palms for a long time instead of falling, sometimes covering the trunk entirely (Plate IIIb). We have ob- served up to twenty dry fronds forming a thick sheath around the stipe. This phenomenoa is more pronounced when environmental conditions are worse. This process reduces transpiration, therefore water consumption and the pelms adapt themselves to a lower metabolic level. Thus, in spite of the length of the droughts, these palms succeed in maintaining a level of existence which enables them to survive until the return of normal con- ditions. Many of them retain all their life the imprint of these periods of extreme privation, as evidenced by a characteristic narrowing of the stipe, which attracts the observer's attention as soon as he arrives in these islands (Plate VIa). III. PALMS WHICH APPEAR TO IGNORE DROUGHTS These may be divided into four groups, as follows: village palms; palms growing at the edge of babai pits; palms growing in abandoned babai pits; palms growing around ponds and small inland lakes. (a) Village palms. In the immediate vicinity of all sizeable centres of population we saw coconut palms, whose production, while 10% high, was still good. Their number is limited, they are in fact confined to a narrow belt around the village, with isolated specimens growing among the huts. Some of these sometime bear heavy clusters of nuts. We saw some which after two years of intense drought bore loads of sixty and even up to a hundred nuts (Tabontebike village on Abemama). In spite of the general cleanliness of Gilbertese villages some debris accumulate near the palms such as that from fish cleaning, shell fish preparation, animal excreta (dogs, chickens, and young pigs not yet kept in sties). The amount of such waste is not large, but is regularly added to. Thus every month, at fixed dates, hundreds and some- times thousands oi fish of the genus Albula (te ikari) are scaled, gutted and washed in the vicinity of the palms (generally always the same ones, from habit or for convenience). Also every month, on regular dates, considerable quantities of te baitari, edible jelly-fish of the genus Tamoya, are cut up and the gelatinous and useless parts buried at the foot of certain palms. Almost every day other organic waste products from the preparation of small fish or shell-fish which are eaten in abundance, make their contribution. It is impossible to assess the quantity, even approximately. To these waste products are added ashes from cooking fires. A few palms also benefit from their proximity to the little enclosures where the natives bathe and profit also from the water drawn from the neighboring well for washing cooking utensils. = 33- (b) (c) (a) Especially privileged palms are those situated in the vicinity of the fresh or slightly brackish water wells. The presence of these wells was the reason.why the natives chose these sites for their villages. They practically never dry up, and even at the end of long drought periods, when their level is very low and the water may have become brackish, the coconut palms draw as much water from the surrounding soil as they can, taking advantage of their exceptional position since digging the well has always necessi- tated breaking up the hard subesoil layer thus creating an area of constant soil moisture. To these factors we must add the important fact that village palms are spaced further apart than the majority of bush palms. Coconut Palms Beside Babai Pits: These palms usually grow on the embankment formed by the earth removed from the pits. Thus they benefit from the mixing of soil layers in a thicker amount of loose soil, and from constant water supply. Coconut Palms in Abandoned Babai Pits: A few rare individuél palms thrive although surrounded ty sterile coconut palms. Generally they are trees growing in old.abandoned babai pits which easily reach the ground water lens, and in addition, profit from the con- siderable accumulation of organic waste in the bottom of the pit and of soil which has slid back into it. Coconut Palms around Inland Ponds: Each of the above categories contain only a relatively small number of trees while numerous coconut palms have the advantage of being in the more or less immediate vicinity of inland water pools which are sometimes quite extensive. For example, ponds ia the region of Nikunau are approximately 1 km. long by 350 to 400 m. wide. During droughts the areas around these pools seem veritable oases, in the midst of general dryness. We were told that these palms are healthier and are productive only during droughts. It may be that, at such times, the water reaches a lower level more favourable to their root system, which is totally submerged when the level is higher. This theory is supported by the fact that coconut palms planted on dykes dividing fish ponds aré extremely healthy (Plate VIb). According to the natives, palms growing on the edges of inland pools where there is always water, provide a kind of reserve for lean years, for their supply of nuts and toddy never fails. It should be stressed that the productivity of these palms is of little importance for copra production. Their use, and it is a valuable one, is to supply the natives of the area with considerable additions to their daily food sup- plies (karewe, moimoto, marai, etc.). They are also an emergency source of supply. -34- PART III: THE COCONUT PALM AND GROUND-WATER We have already pointed out the importance of water in agricultural ecology on the low coral islands, and how the behaviour of the coconut palm there depends to a large extent on the possibilities of water supply from the soil at root level. It is often stated that the coconut palm likes salt, but, from our ob- servations, the productivity of the coconut palm in the Gilbert Islands is higher in the neighbourhood of the fresh-water "lens". This does not change _the fact of its resistance to salinity (up to 2% of chloride, according to Frappa in Baas Becking, 1949). However, it seems that even if the quality of the water available does not matter very much, it must be abundant; which confirms the remarks of Doctor Tammes (in Baas Becking, 1949), for whom water is the factor limiting the cultivation of the coconut paln. In the Gilbert Islands, the most favourable conditions for the coconut palm are deep sandy soil with fresh to brackish water. On the other hand, conditions are unfavourable either when a hard layer prevents root penetra- tion or excessive water remains stagnant at root level. I. Obstacles to Root Penetration Root penetration may be hindered when gravel or sand are cemented at a shallow level. This is particularly noticeable on the rampart, ocean side, on the majority of the islets (see Plate I). These conditions may be less unfavourable than the following. These layers are, indeed, more permeable than the breccia forming the rocky platform. The brittleness of some cemented formations ("te batano" of the natives) may be such as to enable the roots to penetrate to a certain extent. The slightest cracks in in- durated layers are used by the roots to reach the water. That is why palms in better conditions than most are occasionally encountered in such areas. We have already pointed out (see p. 9) that in the centre of the reef islands, the rocky platform was covered only by a thin layer of soil con- sisting mostly of coarse coral debris. These are most unfavourable condi- tions for the palm especially during droughts. In the rainy season a temporary water supply may form here and there over the rocky platform, but palms in such regions always give a mediocre yield. II. Excess of Water at Root Level Remaining Stagnant This may occur on reef islands during periods of normal rainfall. The water supply formed above the shallow bed-rock may remain stagnant at root level and injure the palms which are as sensitive to water-logged conditions as they are to drought. Similar conditions are realized when coconut palms are growing in old babai pits with an impervious floor. Although relatively infrequent, this case is worth mentioning, since on several islands we observed coconut palms ea5e killed through "asphyxia" of the roots caused by water stagnating in the bottom of the old pits in which the trees had been growing productively for several years. Summary of soil and water relationships (i)' Deep sandy soils with fresh to brackish ground-water are best suited to the coconut palm; (ii) A hard or impervious layer hampering root penetration or natural drainage is always more or less harmful to the coconut palm; and (iii) The coconut palm may grow and produce, particularly during droughts, in the immediate neighbourhood of stagnant water, which may be even very brackish, provided that the roots are not in direct contact with the water; that is, when the palm is planted on dykes or knolls. (iv) The relationship between the quality of the soil and the pro- ductivity of the coconut palm has been discussed earlier. As for the.soil itself, the high calcium carbonate content suits the palms but the importance of the organic matter content of the Soil cannot be overestressed. Before we think of using mineral fertilizers, the practical problem must be solved of the supply of organic matter, of which the palms stand in real and urgent need. A special case may be discussed at this point, that of Betio islet, scene of the battle of Tarawa. This was completely devastated during the battle, and has been replanted to coconuts since the end of the fighting. In many places the young palms are developing with remarkable luxuriance. Some only 6 years old were already producing nuts and toddy. This may be attributed to the 3000 bodies buried on this islet and also to the sanitary habits of the Japanese garrison during the occupation, which may have added as much as 40 kg of excreta per month per man for 20 months. There may have been also some benefit gained from the indirect action of nitro-explosives, abandoned scrap iron, etc. Whatever the reason, the ferility of some areas of Betio is probably the best in the archipelago. On the contrary the coco- nut palms planted on the site of an airstrip in the center of the islet are not very flourishing. - PART IV: CULTIVATION OF THE COCONUT PALM Until the present generation, the Gilbertese did not consider the coco- nut as a cash crop; for him, it was only his basic food supply and the gen- erous provider of most of his requirements. Nowadays, although most of the nuts are gathered for copra, the Gilbert Islander still retains his archaic ideas about his palms, and whilst appreciating the advantages his copra may bring, he has not yet acquired the concept of cultivation corresponding with such a valuable commodity. -36- Nuts for Propagation: The natives' choice of coconuts for planting is some- times guided by size, a very "chancy" procedure, or by the fact that the nuts come from a heavily-laden tree, which may be a surer method. This type of tree, known as a niningaun or "the tenfold coconut palm", is distinguished from the others by the abundance of the bunches and the large number of nuts it bears (Plate VIc). Trees grown in or near villages are generally the most productive, and among them one finds the largest number of niningaun. It is from the latter that the native selects the seed-nuts for planting in or near his village. But for "bush" plantings, different conditions obtain, and seed selection usually depends on chance or convenience. If, near the spot where a tree should be replaced or a gap filled, there is a palm which may be con- sidered a good seed-bearer from its outward characteristics - a fortiori a niningaun - it will be chosen by the native in search of seed. If there is no such tree, he will select nuts from any nearby palm. Unfortunately he will usually be content with any nut. Often he will just take some from a heap thrown together during a copra “campaign”, and which, abandoned, have germinated where they lay. One can imagine the proportion of mediocre, deficient or degenerate trees which results from such a practice. Ripe nuts fall to the ground and if they are not collected, they germinate and take root as best they can. This happened with those which were not collected during the war because copra could not be exported, and even today, many nuts are not harvested, either from neglect or because they lie hidden in a thick tangle of vegetation. These nuts increase still fur- ther the proportion of spontaneously-grown coconuts in the bush. Planting: Not only does the native fail to establish nurseries, but he does not even plant in holes in the "plantation" sense of the word. At the most he will cover with a little earth the nut he has chosen. Neither is there any methodical alignment of young seedlings. Nowhere, therefore, will regular spacing be found, with one exception, of which examples are found on all islands where war-time installations involved the destruction of coconut palms. The owners received compensation at the rate of 6 shillings per coconut palm to be replaced in the bush, and 10 shillings per palm to be replaced in areas levelled for airfields, on condition that such planting be really carried out. (The extra 4 shillings were justified by the harder work involved in breaking up the surface of compacted coral and the mass of piled-up metal landing matting on the strips.) In these areas the planting ef young trees was done in lines, generally with correct spacing. But usually the work was limited to breaking the landing strip surface just enough to bury the nuts. The seedlings, therefore, usually had to struggle for existence. In addition, the intense heat on the sun-beaten cement-like surface was all the more harmful as planting was followed by two years of drought. The young unshaded palms were yellow and weak-looking. Subsequent copious rainfall certainly improved their condition, but, because of their careless planting this was not as beneficial to them as to palms replanted in loose soil in the bush. Exceptional Methods of Cultivation: Two original methods are used in special areas on the island of Nikunau. One is the planting of coconut palms on top ete of dykes built by the natives to separate fish-ponds. Their root systens penetrate these earthen dykes, clinging firmly to them, and only the tips of the roots are in contact with soil that is always sodden, even during severe droughts (Plate VIb). Noting the efficiency of such a method which protects the root syste against total immersion the natives carry things a step further. Thus one sees on swampy ground near the Nikunau lakes a still stranger method of planting. It consists in cutting old coconut palm stumps about 20 inches above ground and using them as a base for germinated nuts (Plate VId). As it grows, the seedling pushes its roots into and around this pedestal, en- gulfing it entirely and finally absorbing it. In this way these new coconut palms have successfully been raised to a level high enough to avoid prolonged immersion of their root system. The natives of the region were not, however, in agreement on the future of such palms. Some of them, with great good sense, thought that their root systems would be so exposed for such a height that they would have to be shored up or at least surrounded by a thick layer of dead fronds or compost. Care of the Trees: The seedling is nearly always abandoned to its fate. In pericds of poor rainfall or of drought, it obviously does not have the water- ing which would be so beneficial at the beginning of root growth. Cleaning is very rarely carried ovt, and then only over very limited areas. Side by side with strips of land invaded by the bush until it is difficult to move about, one will find a piece of land roughly cleared or a perfectly clear space. This is very much an individual matter, the ower being free to do as he pleases with his ow ground, even to neglect it entirely. What is true for weeding is equally true for replacement work. When we speak of cleaning we should distinguish between complete clean- ing as carried out in coconut plantations, a principle which could not be rationally applied to the Gilbert Islands, and partial cleaning which con- sists solely of ridding the area of the bulkier obstacles, such as dead tree trunks and thickets of Scaevola, and other plants which are sometimes so numerous and dense as to make movement almost impossible. Clearance by fire must also be discussed. Such a procedure obviously requires no effort but usually has deplorable results. There is no ac- cepted policy for cleaning land in these territories except for that carried out for a width of ten feet or so along the roads running on the lagoonside of every island from one end to the other. These operations, however, are intended merely to clear the road and not to benefit the coconut palms. The method is by no means general and some islands like Aranuka (almost entirely) and Abemama (to a large extent) show almost a complete disregard of this required simple operation. Fertilizers; No fertilizer is given to the coconut palm except for a piece of scrap iron sometimes placed in a hole with a germinated nut. ~ 38 Control of Pests: There is no pest control. Luckily the coconut palm enjoys almost complete freedom from large-scale or intense attacks, so no control measures are necessary in the Gilberts, except on the island of Abemama. There, an orthopteron of the genus Graeffea is abundant and a menacing pest (see p. 28). This phasmid, owing to its abundance, is obviously well-known to the natives of the district, but they are heedless of the damage it inflicts on the palm fronds. We were told that large fires were lit under the trees to make the phasmids fall off, but the truth of this statement is debatable. It is more likely that the insects fall because of the intense heat generated under the trees by fires lighted for cleaning purposes. Gathering and Drying: Nuts for copra are gathered after they have fallen naturally, while those nuts required for food are picked from the trees. Exceptions to this do, however, cocur when occasionally some native, tempted by the arrival of new supplies, needs a little copra to buy sugar, flour, tobacco, or cloth at the store. In current practice nuts are gathered by the owners of the land, and as most of the patches of ground are very limited in area, the quantity of fallen nuts is not very large. A couple, or the wife only, brings the iuvis to the spot where the husking is to be done. Often they are carried tied together two by two by a strip of fibre pulled from the husk. The coconut is opened either on a stake firmly driven into the ground, with a sharply- pointed end upwards, or more often with a little hatchet. A very sharp strong knife is then used to make radiating or parallel incisions in the oily albumen, detaching it from the endocarp. The meat is placed in sacks woven on the spot as needed, from coconut palm fronds (Plate VIIa). It is then taken to the village where it is spread to dry in the sun either on mats (Plate VIIb), on trays made of coconut fronds, on old tarpaulins, or even on old sheets of corrugated iron. The time necessary for drying will obviously depend on the weather. The degree of dryness of the kernel is also a factor. Sometimes this is so advanced that as soon as the nut is opened, the kernel can be detached whole. Once the copra is dry, it is delivered to the Boboti (Cooperative Societies). When harvested in areas remote from the purchasing centre, as cccurs on large islands, it is brought to pre-arranged places (Plate VIIc) for collection by the Co-operative lorry which tenes the sacks to a copra point usually situated in the immediate neighvourhiood of a land- ing stage. Whale boats are used for loading the copra into cargo vessels. It may happen that the native, eager to buy some more or less useful object, will bring not copra but whole nuts, and will sell them at the rate of 3 shillings per 50 nuts. As a rule he stores his nuts in emall covered, latticed enclosures, te okai (Plate VIIa). These crude huts erable him also to keep a smail emergency food stock for perisds of drought, of which undue prolongation is always to be feared. Apart from these comparatively rare cases where the native, in a hurry, sells his nuts whole, the trans- actions are made with dry copra. Commercial Quality: Gilbert Island copra is sundried and of relatively good Sa ES ee quality by South Seas standards. Although there may be abundant rainfall =39- almost daily for several months, accompanied by high humidity, the copra is scarcely affected as the quantities are always small and therefore easily sheltered. Age and life-span of the Coconut Palm in the Gilbert Islands: Generally speaking, we have no dependable information on the age and life-span of the coconut palms in the Gilbert Islands. It is difficult to give any figure, but under the usually inediocre conditions of the Gilbert Islands, a fifty to sixty year old palm may be considered as already old and its yield on the decline. Causes of this premature ageing include irregularity of rain- fall, severe drought, occasionally lasting several years, other physical factors prejudicial to the coconut palm, soil deficiencies of various kinds, lack of any real selection and even of elementary attention to the young palms (bad rooting of seedlings) etc., and finally, overcrowding. As a result of these numerous difficulties - and we have enumerated only the most important = coconut palms here have a relatively short pro- ductive period. They do remain alive for many years beyond this, but have no economic value. It is to be regretted, therefore, that so many unpro- ductive trees should be allowed to occupy valuable space for such a long time. Yield: The only records available are the figures of the 1947 census which gives the number of productive coconut palms for each island. These figures are based on a rough estimate and however interesting, cannot be used in an agricultural survey as they are not the result of counting tree by tree over a given area, which is practically impossible. To accept what the natives say is also misleading, for occasionally there is a 100% difference between their answers and later checking of them in the field. Consequently we prefer to rely on the results of our own calculations. Let us mention first a count made over the whole width of one section of the island of Bikenibeu (Tarawa atoll). The plan in Fig. 14 accurately plots 138 coconut palms of productive age and all young palms grown probably from nuts which germinated where they fell. The position of the pandanus trees is also indicated. This plan should be studied with Table V which gives the number of fronds, green nuts and ripe nuts, counted on three dif- ferent dates. This site was chosen for our inventory because it appeared to us to represent an area of average density of palms for this atoll. We counted 138 palms over an area of 5,950 sq. metres or 231 per hectare, but it should be noted that the spacing of these trees is very irregular. The first inventory of these 138 numbered palms was made on the 2lst March, just before the beginning of the regular rains which lasted several months. It resulted in the following averages per tree: 19 leaves, 4.86 green nuts, 0.56 ripe nuts. The second inventory made on 6th June, seventy-six days later, gave an ESS of 22 leaves, 10 green nuts, 1.6 ripe. The third in- ventory made on Sth August, or exactly two months later, finally gave 24 leaves and 13.6 green nuts (the ripe nuts had been gathered by the owners). -0- TABLE V Observations made at BIKENIBEU (TARAWA) on 133 coconut palms on an area of 5950 Sq. meters. (Rate of 231 trees per hectare. ) 21 March 6 June 6 August * No. Of No.. oF No. of Wo. of No. of No. of Tree No. of Green Ripe No. of Green Ripe No. of Green Ripe No. Leaves Nuts Nuts Leaves Nuts Nuts Leaves Nuts Nuts 1 22 7 4 28 14 1 3h 19 2 19 a ‘ 21 8 ; 22 9 3 17 ; ; 23 : 26 2 4 10 12 6 13 8 5 16 20 3 20 2 6 7 3 al 12 9 12 11 ‘ 7 3 ; 9 6 10 2 8 7 4 12 x ) 13 9 7 : : 1 3 12 7 10 9 2 : 14 13 : 16 19 11 20 10 4 ek 22 ‘ 26 27 12 6 ; 3 9 i i 1g 29 13 20 3 3 ah, ful Q 26 12 14 18 3 : 22 10 ‘ 25 13 5 20 8 h 25 9 8 27 32 16 21 , : 26 ‘ ek 3 LT 16 4 ‘ 20 2 h ea h 18 13 ; , 15 2 . 19 3 19 12 ; 16 20 : 16 22 20 16 20 4 2 22 13 el 16 h iF 20 19 Th 26 25 22 15 6 3 18 14 6 19 12 23 14 A i 16 : 18 2 ek 14 d 3 18 10 d 2k 19 25 UT 3 h 20 30 2 27 28 26 15 7 3 18 y 5 18 5 27 10 9 ; 14 10 a aly 15 28 14 2 3 19 6 3 19 10 29 29 6 : 12 ; 13 30 19 H 34 6 : 36 23 en 21 7 3 ek i 25 3 32 27 7 25 el i 31 LO 33 al 10 23 10 5 20 13 34 20 : 2 26 2 2 22 35 a7, 6 20 3 23 2 TABLE V (continued) 21 March 6 June 6 August * No.of Nowvor NOs 280k » NO OL No. of Tree No. of Green Ripe No. of Green Ripe | NO. Of. Green No. Leaves Nuts Nuts Leaves Nuts Nuts Leaves Nuts 37 18 a ’ 23 16 h oe 30 38 18 : : 22 7 A 21 11 39 ok 14 2 28 14 10 29 Te) fe) chee Eas esis x 21 31 2 28 20 Ta 19 6 : 23 20 h 27 22 he "13 j : 16 6 f 19 6 43 17 6 , 22 Ou: 3 22 4 hh 23 : 5 27 : F 28 15 ks 18 13 : 22 : 10 20 46 10 : 14 é ’ 20 1 LT 24 , ' 27 é 34 60 48 21 13 ; 2h 53. 27 11 Ke) 18 7 : 21 i 26 5 50 15 i i 18 19 22 1 51 21 : ; au : a 26 2 52 13 i : 16 1 1 18 27 53 16 10 7 20 26 10 ay 2 5h 27 : 1 29 2 29 2 55 20 3 i 23 1 Rb 23 2 56 15 3 : 18 h 3 21 6 57 2h 7 27 : : 29 “52 58 25 21 28 ko ly 4 “31 AFL y. 20 13 au 67 1 28 27 60 27 12 30 LO 32 19 61 28 32 38 70 62 27 17 an 66 32 2 63 15 16 18 2 20 7 64 23 4 26 5 2 26 26 65 ek 7 26 20 3 29 a5 66 22 : 25 5 ; oh 12 67 25 i 28 vf , 26 8 68 25 6 29 14 ‘ 27 16 69 22 if 2 26 a } 28 36 70 32 a7, 2 37 31 iM hi rel 21 2 2 eh : 2h 72 21 3 25 3 6 23 22 fs: 20 4 au 12 5 26 19 74 23 26 i il 27 iE. Al | 20 ‘ 19 76 6 8 8 ai 22 13 25 25 13 2 25 TABLE V (continued) 21 March 6 June 6 August * No. of No. of No. of No. of No. of Tree No. of Green Ripe No. of . Green Ripe No. of Green No. Leaves Nuts Nuts Leaves Nuts Nuts Leaves Nuts 78 19 na) y 22 22 2 21 : 79 20 a7 ; 24 13 ‘ 26 10 80 21 3 H 13 k ; iUe. ‘ 81 a7 2 iL 20 ; a 26 2 82 ek 12 : 26 22 ; 25 20 83 _ 20 8 : 23 8 7 22 1 8h 6 ; ; 6 ‘ 9 85 20 10 : 20 16 6 ou i 86 16 17 s 18 22 4 22 19 87 19 4 19 : 22 88 27 30 8 30 89 26 7 28 8 3 30 eh 90 26 h 28 1 28 : 91 22 22 1 22 : 92 19 fi 21 1 26 i 93 20 il aS 28 5 4 2 9 94 21 11 3 an h 5 26 6 95 23 7 al 27 12 5 28 if 96 16 : 19 5 14 A 97 20 ib 23 23 an 9 98 22 26 27 3 99 30 7 4 3h 26 5 Su 33 100 21 Lg 26 54. 28 99 101 21 22 23 102 3 5 aha 103 16 12 20 L8 22 55 104 28 17 31 hh i 34 32 105 22 | 26 1 2T 3 106 22 ul 2k 18 2 eh 26 107 23 17 26 37 27 Ufo) 108 19 7 21 10 22 6 109 20 22 15 26 23 110 20 7 23; 28 30 33 a 26 ab. 33 20 a 31 16 112 30 13 30 i 30 113 27 3 27 3 26 114 el ! 23 2 23 115 13 16 7 21 116 20 ! : 26 : A 29 f slay) 23 k ! 27 1: : 29 18 118 23 ! 26 12 18 ; 119 30 34. 30 TABLE V (continued) 21 March - 6 June 6 August * NO: Or ~~ GO. a Noy of "Nol or No. of Tree No. of Green Ripe No. of Green Ripe No.of ~ Green No. Leaves Nut Nuts Leaves Nuts Nuts Leaves Nuts 120 22 Ly ; 25 10 al 30 18 121 28 : it 30 : : 31 8 122 12 i 2 14 8 5 16 2 123 20 : ‘ 22 5) ij 23 124 9 10 ‘ : 12 125 26 29 29 126 22 oh 5 ek 127 22 26 h Zl 2 128 20 23 16 St 19 129 22 9 3 27 20 31 8 130 33 2 37 6 26 131 27 11 29 28 33 38 132 31 2 34 10 28 26 133 18 a 22 27 3 31 18 134 26 11 3 30 16 5 34 11 135 10 10 16 136 26 30 32 53 37 37 ho 6 0) 10 5 138 el 22 19 Total 2694 671 78 3124 1454 221 3315 1887 Average: 19.52 466, O,56 22.63 20254. 1560 24.02 13.67 * On August 6 there were no ripe nuts. No. OO ON OW FWwWNEH TABLE VI Dimensions and weights of 50 coconuts at Bikenibeu, Tarawa. Nuts from recorded trees. Length (Inches ) Jil ee 9 8. ) = ome) be MO OOOF © roo CW WOW OW WM WOM UID MW Www WIV) WI WU) 2 5) Wh Wt WNT Circun- ference (Inches ) ) Whole Weight lbs. PREP HORORPNPHEPBRBPRPRFPEPNNFPEFPRFORRPREPEPHPHERPEFEFHOFRNOHFEF OZ. 12 L a) ) Y) 1) ~ r PMMO FO ORO FFOOOOFRPANOGNFPNOCNN OF OO j Weight with- out water lbs. PRE HHOKOHNRFPORPOPRPENEPHRPRPRFPORPHEPEPRPEPEPRPRPEPRPFOFRPFHFOFRFEF OZ. 12 = OF OMN OV he ee PN FOFO OF OWWUAFOWH AU GAH NDOOAFE ee Weight of fresh meats OZ. WUOIADAWO FAOANNEHNAWMO FOUANDNEFAHNOU AWN OOO! Ov] © TABLE VI, Cont'd. Circun- Whole Weight with- Weight of No. Length ference Weight out water fresh meats (Inches ) (Inches ) lbs og ibe: OZ. oz. ho 8 14 ib 10 1 0) 6 h3 10 18 0 13 6 = - hh 10.5 14.5 1 he fy sinc 6 5 o> 15 a 12 i Me 8 46 8.5 aye ie 14 1 TEs) 6 LT 10 V6E5 aly 5 aL h fi 48 10 iS a 8 1 1 5 lg 10 1625 x6) 12 fe) 12 3 50 OES 16 it (0) aft O Hay bs 18 lbs 7 o2 Total weight of fresh meats 418 lbs. 7 -az. Total weight of dried meats (copra) 14 lbs. Average weight copra per nut 0.281 lbs. Loss due to drying 25.9 per cent Nuts required per ton of copra 7874 TABLE VII Dimensions and weights of 50 coconuts taken at random from an Abemama plantation. The trees are regularly spaced, about 234 per hectare, and show outstanding vigor. Circun- Whole Weight with- Weight of No. Length ference weight out water fresh meats (Inches ) (Inches) lbs. oz. ibs. OZ. oz. i 12 19 i ~ 4 2 1.5 eS ye 12 ii au 9 3 aay 20 ai 10 - 8 Ae 9.5 oy koe @) 7 5 10 £325) 2 h al 3 6 6 9 PS e5 0H 7 1 by W if 10 LF Sie 6 - 6 8 12 14 aL 3 - 7 9 cy 15 1 12 i 10 7 10 10 18 1 11 i 10 12 aa 9.5 18 i 8 - 9 12 as 18 2 i - 10 13 8 10 ik 2 = if 14 12 21 2 2 ~ 10 15 12 eS eat 12 = 8 16 10 15 2 7 - 8 17 8 a aL: 6 1 5 Ul 18 1. 5 18 a 15 - 10 19 1055 21 fl 12 13 20 Mae iG 1 2 - i 21 i) 20 1 15 a 13 8 22 10.5 15 il 4 = if 23 8.5 US eel ee © 6 eh 10 15 a 9 a it 8 25 1255 aS alt T i 6 if 26 We 16 1 12 1 7 7 27 10 13 1 2 - 6 28 10 LS 1 3 - 8 29 OES 15 iL 12 1 6 6 30 9.5 16 1 h a 3 7 oi 10.5 EW ets ite Wil 5 8 32 9 18 1 2 7 33 10 TUS ie 3 1 fl 8 3. 9.5 Ly) 1 rf i y 7 35 9.5 16 1 2 - A 36 9.5 15 1 6 nu h 6 3 SB, ik 1 8 i 7 it 38 10 17 - 14 - 6 39 8.5 BMS es YL 4 i 2 6 Lo 10 14 2 a al 12 8 TABLE VII, cont'd. Circum- Whole Weight with- Weight of No Length ference weight out water fresh meats (Inches) (Inches) Ibsuuaneze lbs. OZ. OZ. hi 2A. ay, ah 3 4 ho 10 18 iE 9) 5 43 LO als) a (9) 6 Ly 10 mB A 4. @) - 6 U5 9 16.5 (6) 14 6 46 8 12 a) 12 4 47 Gia TESS a 2 4 48 oe 14.5 Ab 2 h Ite) 8 12 ab 3 6 50 8.5 12 (@) 14 5 Total weight fresh meats 5126) dhs: Total weight dried meats (copra) 16.88" tps, Loss due to drying Average weight copra per nut Nuts required per ton of copra ey 25.0 per cent 0.335 lbs. 6578 TABLE VIII Dimensions and weights of 25 coconuts collected at random on the Reiher property at Abemama. The trees of the plantation are spaced 8 by 8 meters. Circim- Whole Weight with- Weight of No. Length fer2-nce weight out water fresh meats (inches) (Inztes)* lbs. ozs lbs OZ. lbs. OZ. lg 10.5 2h 2 8 - : ee 0 2 10 19 2 0 - 6) ve 3 10.5 20.5 a 14 - ) 10 h Aa 19.5 a 8 - 0 it 5. LOSS 19 qi 9 = ¢) 10 6 10.5 14.5 ) 5 = O 6 7 re aM Bi 9 = 6) 6 8 41. 19.5 1 8 - 6) 7 9 9.5 14 6) a - 0 p 10 9 14 1 h ab 3 ) 8 11 9 14 8) ad - ) 6 12 10 BS5 a 0 ° 6) 6 13 9 155 il 0 - 0 6 14 9.5 14 @) 14 ~ O 5 15 10.5 1555 a 4 - ) if 16 10.5 6a5 1 14 a, 13 0 8 17 9.5 14 @) 15 - 0 6 18 eS 18 z 8 = 0 t 19 9.5 17 1 3 - 0 T 20 aS 20 aL 12 “ 6) 7 al 10.5 20.5 a 13 - fe) 8 22 9 AL 1 ih - ) 9 23 12 20 ip 12 - @) 8 24. LO 85 20 1 15 ik 14 @) 12 25 11 18 iL 4 - 0) 8 Total weight of fresh meats 12,31 ibs. Total weight of dried meats (copra) 8.56 lbs. Loss due to drying 30.00 per cent Average weight copra per nut 0.342 lbs. Nuts required per ton of copra 64.36 TABLE IX Size and weights of 25 coconuts from a single superior palm on the Reiher property at Abemama. The trees of the plantation are spaced 8 x 8 meters. Circum- Whole Weight with- Weight of No. Length ference Weight out water fresh meats (Inches) (Inches). lbs. oz. lbs. OZ. lbs. OZ. 1 10 19 i a - @) 8 2 8 18 Hl 2 = 0 Hf 3 6.5 18 i 8 - @) 9 4 9 20 2 8 « af fe) 5 9 18.5 ab if = ) 9 6 9.5 19 a 8 uM if 0 13 i 9 19 al 10 - @) 14 8 10 el ik 12 > 9) 10 9g 2 19.5 1 6 = 0 9 10 9 19.5 ae 7 - @) 8 tal: 9 19 uh 10 ug 3 9) 9 WW 9.5 19.5 1 9 - 0 8 13 8.5 18.5 n 10 1 8 @) 9 14 9.5 19 ab 6 - ) 7 15 Si) 18 a 6 - ) t 16 9 19.5 ai 6 - fe) 7 aly 9.5 20 al 8 - 1) 6 18 9 19 1 9 i. u ) 8 19 9 ie 5 af "4 a 6 0 8 20 9 19 1 5 = ) 7 ene 9.5 1355 a 10 i 8 9) 8 22 8.5 19.5 al 8 i 1S @) 9 eg 10 20 ib 8 = 0 8 2h 9.5 20 a 10 - 0 9 25 9 20RD 2 10 - 0 LE Total weight of fresh meats 14,0 lbs. Total weight of dried meats (copra) 10.78 lbs. Loss due to drying 21.42 per cent Average weight copra per nut O.440 lbs. Nuts required per ton of copra 5000 Among other things, these figures show the rapid response of these palms to the first heavy rainfall after two years of severe drought. Table V also shows that a number of very old or unproductive palms should be removed. The usefulness of continuing such observation work for several years, during normal rainfall and drought, can readily be appreciated. Valuable conclusions can be based only upon data gathered for a sufficiently lengthy period of time. Secondly, let us mention the coconut palins on village sites. Their density is always very low by comparison with palms growing in other areas of equal value. Fourteen counts made in populated areas, gave a general average of 115 palms per hectare. Let us take for example the village of Tabontebike where, for an area of 10,750 sq. metres we counted 100 pro- ductive palms representing 93.02 per hectare (plus 36 Artocarpus). These palms, had, on the average, 17.34 nuts when the count was made before the rains, This is not an exceptional number in such conditions after prolonged drought. Thirdly, we shall mention the figures from Makin School (Abemama). The area was chosen because of the regular spacing and vigorous appearance of the palms. We counted 23!: palins per hectare with an average of 18.15 nuts (drought production). It should not be inferred frm these figures and averages that there is any absolute regularity in spacing (with the exception of the Reiher planta- tion on Abemama), the density varies constantly even for one hectare; for Village palms it will vary from 80 to 150 and, in bush regions, from 200 to 350. The map of Bikenibeu (fig. 14) shows the irregular spacing of the palms. With regard to the weight of ripe nuts during droughts, Table VI gives measurements and weights of 50 ripe nuts from Bikenibeu (Tarawa). These come from some of the 138 numbered palms (fig. 14, table V). The length measurement was made by following a ridge from the distal extremity to the centre of the point of attachment; the circwaference measurement was made where the diameter was greatest. The nut was first weighed whole and then again when it had been opened and emptied of its liquid, if any re- mained. The fresh meat was extracted and weighed separately for each nut. The copra obtained after four or five days! drying was weighed. We thus have, for the region of Bikenibeu, after weighing 50 apetg an average per nut of 0.127 kg. of copra, the loss in drying being 25.9%. The number of nuts per ton of copra is, therefore, 7,874. Similar weighings were carried out on several islands. We shall mention three for Abemama (centre group ) and one for Nikunau (southern group). On Abemama we chose the palms of the Makin School because of their healthy growth and good production and also because, while not planted with equal spacing as on a real plantation, they were fairly evenly dis- tributed. We counted 234 palis per hectare and obtained an average per nut of 0.152 kg of copra, with a 25% loss in drying (Table VII). ee Ny le On Bikenibeu, we had practically the same number of palms per hectare (231), but they were very irregularly spaced, varying from 1 to 8 per 100 sq. m., whereas those of the Makin School never exceeded 3 per sq. m. It is also important to remember that at Makin School the width of the island is ap- proximately 700 m, whereas on Bikenibeu it is only 200 m a difference which certainly influences productivity. Two other weighings from Abemama were made on the Reiher property where palms actually planted in the agricultural sense of the term and grown from selected seedlings (mass) are spaced at intervals of 3m. The first of these weighings (Table VIII) was of 25 ripe nuts from different trees, all showing distinct signs of drought. These nuts gave an average of 0.155 ke of copra per nut, with a 30% loss on drying. ‘The second of these weighings (Table IX) was for 25 ripe nuts from a single palin growing beside an old babai pit, and therefore in a situation where water would be less likely to fail. There we had an average of 0.200 kg of copra per nut with a 21.42% loss in drying. It is worthy of mention that this remarkable tree had 34 fronds and 93 nuts, of which 29 were ripe and provided the 25 we examined. Considering this production, and the great leaf development, we may conclude that the averages for 25 nuts give a fairly good idea of the qual- ity of the yield of this plantation in a normal season, as the selected tree enjoyed very favourable conditions in spite of the drought. According to this last weighing, we calculated that 5,000 nuts were needed to obtain one ton of copra. It is difficult to base final general conclusions on specific cases; and our long experience of plantation production statistics taught us to be prudent in this matter. At the time of our stay on the Reiher property (April) the effects of rain had not yet appeared and 90% of the palms showed traces of the very severe drought which had lasted for two years. But the size of the stipes was evidence of the possible vigour of these trees in normal times: so also was the total number of their leaves, including those which, as a result of the He Soe were hanging down the trunk (often more than twenty, Plate IIIb). Another similar survey was made during our stay on the south islands. On Nikunau we bought 50 nuts at random and dealt with them in the same way as with the previous groups. These nuts (Plate VIIIa and b) were rela- tively tiny (three or four times smaller than on the north and central Gilbert Islands) but were all perfectly formed and their copra was as good in quality as that of normal nuts. For these 50 "drought" nuts, we got an average weight per nut of 0.051 kg of copra, with 36.35% loss in drying. The number of such nuts required to produce one ton of copra was here 19,380. Table IV lists sizes and weights of these 50 Nikunau nuts. Estimated Yield on Tarawa: For the following estimates we worked from figures in the "Report on the 1947 Census" by Mr. F. N. M, Pusinelli, for the area, population and number of animals (pigs) eating coconuts. For estimates of density of trees and production of coconuts we shall use our own calculations based on the counts of trees and weighings of nuts which «jd. we carried out in the field, and the averages thus obtained. Obviously, these figures are approximations, in the absence of more accurate sources. The area of Tarawa atoll is estimated to be about 2,000 hectares. It is considered that 20% should be deducted for uncultivated areas such as roads, swamps and mangrove. The area actually occupied by the coconut palm is therefore 1,6000 hectares. If we assume that 231 trees per hectare on Bikenibeu give a fair idea of the average coconut palm figures for the whole of the Tarawa atoll, we obtain 369,600 palms for the whole productive area. We must deduct from this figure about two palms per group of five people which correspond to the number used for toddy, namely 955 trees, which leaves 368,645 palms as nut producers. Pusinelli's census estimates that, of the total population of Tarawa (3,582), about a third are not natives of this atoll but come from outside to work at "colony headquarters". We thus have a figure of 2,388 persons considered to be really natives of Tarawa and consequently owners of land. We estimate that the average consumption for ea 1) individual is four nuts per day, which gives a figure of 3,486,000 nuts\l), We also calculate, from the figure of 768 pigs in the 1947 census, that read took of nuts by these animals, at a rate of three per day, is 841,000 nuts per year. Finally, considering that the export of copra for 1949-1950 was 662 tons and that the average of copra per nut (compare Tables VI, VII, VIII and IX) is, for the central Gilberts, 158 g, these 662 tons represent 4,190,000 nuts. We thus have: Annual consumption (population) 3,486,000 nuts Annual consumption (pigs) 841,000 " Export of 662 tons of copra EMIOO SOCOn. 0. 8,517,000 " In conclusion, if we divide this number by the number of nut-producing palins (368,645), we obtain 23.1 nuts, representing approximately the annual average production per paln. The nuts destroyed by rats and crabs, fed to chickens or sold to non-Tarawans are not included. Their number is probably not large enough to be significant. (T) Dr. Catala'ts estimate of consumption is an under-estimate; whereas it is true that one-third of the population is not native of Tarawa, that one- third eats a great deal of nuts. Furthermore, no account is taken of nuts which fall and are not harvested - this is thought to be a fairly large quantity. (Note by the Administration of the Gilbert and Ellice Islands Colony. ) -43- We have adopted a minimum figure for nuts consumed by pigs, for the number of pigs in 1949-1950 must have been substantially lower than in 1947, the census year, which had a normal rainfall. Indeed, when the Gilbertese sees that a drought is going to be unduly prolonged, he reduces the number of his pigs. PART V: USES OF THE COCONUT PALM FOOD Te Karewe - "Toddy" The production of toddy from the coconut or from other palms is prac- ticed in several parts of the world, but nowhere has it such fundamental importance as in the Gilbert Islands. Karewe is drawn or tapped from the unopened inflorescence after a series of manipulations requiring a precise technique (Plate VIIIc and d). When a new spathe appears and the native has observed that it is half a hand in length, he will allow approximately twenty-eight days to elapse before treating it, but some toddy experts indicated that there is an ad- vantage in choosing such an inflorescence at the new moon and regarding it as ready for use at the full moon. The spathe, which meanwhile has reached its full length has not yet split. The first operation consists in binding it with a coir string to keep it from opening up. The elements of the in- florescence will be unable to spread or open. The inflorescence then will be freed at the extremity for a length of about 10 cm, from the spathe pro- tecting it. A very tight binding will replace the spathe to keep the flowering branches from spreading. This mass which is originally erect is gradually bent with a rope tied to the petiole of a lower frond (Plate VIIIc). This slow curving takes about 3 days, but varies with the technique of the operator and may take less time. After this time, the extremity of the inflorescence is cut off with a sharp blade in thin slices, 3 times a day for 3 or 4 days. The liquid, which exudes in small quantity at first, is not collected because it is too tart. This preliminary phase over, the first container, an empty coconut shell (Plate VIIId) or glass bottle is attached to the stump. A fragment from a nearby leaf is also attached to it to guide the liquid into the container, This fragment must include the midrib, a small sec- tion of which is freed from the blade and used as a pin to hold the frag- ment in place. The abundance and rapidity of formation of the liquid varies with several factors: quality of the individual tree, time of production, technique of binding and especially, according to the experts, degree of binding. Finally the precision with which the wound is reopened is also very important. This is done morning and evening when the filled containers are fetched and replaced with empty ones. Rarely do trees produce enough to need renewing of the cut in the middle of the day. Every time the operator reaches by repeated slicing the part of the inflorescence still covered by the spathe, this must be shortened again. -4he Toward the end of the process when the spathe is only about 15 cm long, it is completely removed. The duration of the exudation is about 6 weeks with about 2 full nut shells filled every 2: hours. The capacity of these containers is about 850 cc. Even the experts could not tell whether the yield was greater at one particular time of the operation. Some inflorescences produce more at the beginning, for others it is the opposite. The choice of coconut palms for toddy production is random, except perhaps in the case of some experts with a special flair, or keen observa- tion powers. Any inflorescence will be used without necessarily waiting for the new moon, and it is only after tapping that it is known whether one has a "winner". There are poor trees and poor tappers, so failure may be due to poor technique or to the palm itself. It is not uncommon to tap the same palm for 15 or 20 years if it is a good producer, but it will be left to rest from time to time when a notice- able decrease in the size of inflorescences is observed. Later on, after a period varying from 8 to 10 months, the new inflorescences will be used for | toddy if they are large enough. As a certain number of palms is always available to each owner, some will be producing toddy, the others nuts. This is, of course, during years of normal rainfall; this alternating use cannot be continued during prolonged droughts, especially in the southern islands, but the inflorescences though unable to set fruit, will almost al- ways be suitable for toddy. This is a fact of the utmost importance to assure the Gilbert Islander of a balanced diet during periods of food short- age. For their supply of karewe, the old people and very young children depend on the strong man of the family for whom climbing the tall coconut palms twice daily is an easy task. If no such man can be found in the family, and no arrangement can be made with a neighbour, a very serious problem arises that is not easy to solve. A Gilbertese will never refuse the loan of a palm, but will very seldom collect fresh toddy for any but his own family. Toddy is never sold. We noted the following as an indication of the average quantity of toddy available to a Gilbertese family: on Nikunau, in the south, our young informant, Kiritama, used five inflorescences producing an average of seven coconut shells almost full, or approximately five litres of toddy every 24 hours. Toddy is generally tapped from only one inflorescence per palm, but during periods of normal climatic conditions and in areas where coconut palms are very good producers, two inflorescences will often be seen in simultaneous production. Three inflorescences are very seldom used on one palm. Coconut palms sometimes are observed producing both toddy and nuts. The flavour of toddy does not vary according to the variety of the palms, but it changes from palm to paln. vie For consumption as daily beverage, toddy is brought home and immedi- ately diluted with an equal volume of water. Unless it is boiled, fermen- tation will be rapid. After 15 or 18 hours, it is already sour. This sour toddy, or te manging, is not often consumed as its after-effects are harmful, a and most eeeirea are wary of it. Its consumption is controlled by license. In the preparation of kamaimai - fresh toddy is boiled until a syrup is obtained. This may be more or less concentrated, and the higher the con» centration the better it will keep. When this syrup is intended for long storage, it is reduced in the ratio of 4 to 1. It may later be consumed, diluted in water. It will also be included as an ingredient of many dishes. Finally, in households where money is not plentiful enough to buy sugar, kamaimai will be used instead. We have said that karewe was never sold. On the contrary, kamainai is sometimes an item for small local transactions. We saw it sold in the Arorae store at 9 pence a bottle. The price may rise to 1 shilling. Preparation called kareberebe - this is obtained by cooking kamaimai until it is completely caramelized, and the resulting product is divided into small balls which make excellent sweets, similar to toffee. Fermented toddy is used as leavening in some kinds of pastries such as the fritters called te tonati. Food value of toddy; Fresh toddy contains 16% sugar, fermented toddy 6% alcohol. Analyses indicate the presence of Vitamin B in fresh toddy. But the high content of Group B vitamins is chiefly noticeable in fermented toddy, this being approximately equivalent to one-third of that found in Tes s yeast (Buchanan, 1947). According to Dr. Bray (in Turbott, 1949, . 40), the protein content of toddy would be 0.32%, the carbohydrate cnn- ae 13.0% with 54 calories per 100 g. Other foods from coconut: The mesocarp (kora) of te bunia variety may be eaten from the stage te ubu until te amaere (see Fig. 15 and Plate IXa). It is chewed somewhat like sugar-cane. The oily albumen is consumed from the gelatinous stage until it has reached complete maturity. At the early stage of marai the jelly-like meat is reserved for infants and toothless old people. Te moimoto represents the age at which green nuts are preferably picked. At this stage the nuts contain a maximum amount of water (te ranin), and the meat is fully developed but still soft. Its Vitamin B content is low, but the water is said to be a good antiscorbutic. The mature nut, ben, is an important souce of fat. It is rich in protein and carbohydrates. The Vitamin A content is negligible whatever the degree of maturity. Grated coconut te ota and the cream which may be extracted from it are poor in vitamins. But the developed germ to bebe is rich in Vitamin C. This sweet spongy mass is much appreciated and mostly consumed by natives while they are cutting copra. -46- To sum up, all these edible parts of the coconut palm have a very low vitamin content. There iiay be a compensating effect in the fact that the natives consume them frequently and in large quantities. These elements are, in fact, included in all food preparations. It should be mentioned that very few Gilbert Islanders eat the palm cabbage even when trees have fallen down or have to be felled. Pharmacopoeia; The mesocarp of the te bunia variety is used at the te ubu, te ra and te moimoto stages to treat cases of infantile diarrhoea (te banibong) in patients 13 to 4 years old. The liquid obtained by beating and squeezing is mixed with coconut water to sweeten the dose. The same preparation is used as a toothpaste. The inflorescence is used to treat gingivitis: nine male and three female flowers are taken before the spathe is open. The tongo (Rhizophora) are grated after being peeled. The whole is mixed and the juice is squeezed and applied on the gums. In the case of very small children the mother takes some of the mixture on her finger and massages the guns. The same medicine in a more diluted form is used as a gargle to treat sore throats, etc. BUILDING MATERIALS AND OTHER USES Trunks: These are used as beams in the construction of all buildings from the small shelters for storing coconuts (te okai) to the huge maneaba (con- munity house), and for building light bridges, pig pens, etc. The timber is also favoured for canoe building. Whole fronds: (Te bani or ba-n-te ni). These are used for roofs, partitions, rough mats (Plate IXb). Rolled up and tied, they form the best torches for night-fishing. Parts of fronds; These are used to make baskets for carrying leaves and com- post to the babai pits and, generally carrying bags of all sizes (Plate IXc). Midrib of the fronds (Te taboa): Their stiffness makes it possible to utilize them not only for partitions in European houses but also floors in native huts. These floors are supported by beams and covered with mats. The midribs are also used to make trays for drying fish and the various pastes and cakes made from bread-fruit, pandanus, etc. Main vein of leaflets (Te noka): These are used to fix pandanus leaves to their supports when a roof is being assembled. Leaflets: The tender white leaflets are often employed for ornamental pur- poses during feasts, but their greatest use is in the making of riri (grass skirts), hats, rather fine mats and other handicraft products. Old leaflets are used for compost and in baskets for babai. -47- Coconut husks (Te ewanin): Beside being employed as fertilizers for Arto- carpus, Cyrtosperma, etc. they are used chiefly to make coir (te benu). After being extracted and retted in fish ponds or in the sea or even in the muddy sand near the mangrove, the fibres are hainimered with a mallet, rough- ly separated and made into string (te kora) by the women who roll it on their thighs. This coir plays a considerable part in native life, since it is used to bind all component parts in building houses and canoes. Cord and rope everywhere replace nails, tenons, pegs and mortises. The husk of the te ni variety is used for coir, at the moimoto and to amaere stages only, the latter being the better. In the Gilbert Islands, coir making for ex- port has not yet been undertaken, but such a development is envisaged for the future. Coconut shells (Te nana): When hollowed out, these form the usual con- tainers for toddy collecting (te ibu). Cut in halves, they are used as cups, small bowls, ladles, etc. Burnt, they provide the best of charcoal. Nuts without kernel (Te wae): These aspermous nuts often attain the same size as normal nuts. When they are barely developed and the fibres of the mesocarp are still compact, the natives use them to make corks. -48. CHAFTER 5 THE PANDANUS Pandanus is a tree of medium dimensions, with slender branches irreg- ularly arranged. The base of the trunk, and frequently also the main branches close to the ground, have long adventitious roots, which are more or less hard. Long stran-like leaves are spirally inserted at the end of branches and have sharp marginal serrations. The dead leaves remain at- tached, hanging. The drupaceous fruit cluster is large, sometimes weighing as much as several kilos. The drupes, which are at first green and tightly packed, grow apart as they mature and when ripe display an orange tinge at their basal, less lignified end. Taxonomy: The classification of pandanus is not clear yet and some authors classify as species what others consider to be simple varieties. Thus D. Anderson (in Stone, 1951b, p. 17) thinks that all the pandanus on the island of Arno (Marshall Group) are simply varieties of the tectorius species, and he enumerates sixteen. The botanist Fapy (1951) also seems to consider that practically all the pandanus in French Oceania should be referred to the Pandanus tectorius of Solander. The shape of the fruits, and more particularly of the drupes, are better criteria for classification than those provided by the leaves which vary in shape and length at different ages. Difficulties encountered in the preservation of the flowers and the desiccation of the fruits of the pandanus, as well as their bulk, do not facilitate research by botanists. The remarks made by Papy on the Pandanus tectorius of coral islands generally agree with ours: "Pandanus tectorius Solander is, in these regions, more widely distributed geographically than the coconut palm, its edaphic requirements being practically nil. The syncarp of the pandanus was used in the past to provide the main food supply for these coral islands. The basal part of the mesocarp of each drupe is less lignified than the rest, and contains a good proportion of starch and dextrose. But fibres and crystals of calcium oxalate sometimes make it unpleasant eating." The varieties which, according to the natives, irritate the lips, are te aramboia in Tarawa, and te irikiri and te irineiaro in Nikunau. History: It is logical to think that the presence of the pandanus in the Gil- bert Islands, as in other regions, dates from long before that of the coconut palm. Many authors believe that the germinative properties of the pandanus seeds are preserved even after long periods in sea water, whereas those of the coconut are more rapidly lost in the same conditions. From this it may be deduced that the pandanus did not, like the coconut palm, -h9- need human intervention and that the first of them which colonized these islands grew from seeds which drifted on to their shores. Later, some varieties - as the meaning of some native terms occasionally indicates - must have been imported, some even well after the introduction of the co- conut palm. In the course of the many migrations which these territories have seen slips of the favourite varieties probably continued to be trans-~ ported as limited transport from island to island still takes place today. This would seem to be the meaning of the name “pandaius people" conferred by Grimble (1933-34) on the Gilbert Islanders. The first populations of these territories must have used pandanus as a basic plant food, until the number of coconut palms in production was surficient to supply them with new resources. Then the pandanus was gradually relegated to a secondary role. Recently, since the coconut palm has gained commercial value because of copra, the native tends to neglect more and more his ancestral te kaina which is thus relegated to the poorest land. Reliable observers (Sabatier, 1939, p. 34) whose presence in these territories over many decades has qualified them to make useful comparisons, have already drawn attention to the notable decline of pandanus on several of the Gilbert Islands, Makin and Butaritari among others. This disuse is regrettable because of the food value of the pandanus, and particularly on account of its many domestic uses. It should be pointed out that a major factor during the last decade has been the large number of trees which were felled for the United States Armed Forces and later for the Government. On many islands a limit on the use of pandanus for building has already been reached. Varieties: Grimble (1933-34) estimates that the natives number more than 160 kinds of pandanus. Bingham (1908) lists 170 different names (although some of them seem to correspond to parts of the plant). Finally, Sabatier reports that Frere Eloi counted 194 different pandanus. Only an ethnologist or a linguist would be interested in the translation of so many words, many of which, no longer have any very distinct meaning for the natives. Some trees, which appear to have certain particular characteristics in common, have different names on different islands. Further, on the same atoll a game used in the north will be either unknown or unused in the south, and vice versa. We must bear in mind that the behaviour of pandanus varies according to its edaphic situation. The physico-chemical nature of the soil and sub- soil, humidity, salinity, etc., are factors which influence the aspect of the trees (height, diameter, leaf development, and perhaps even size of fruit). Pandanus grown from seed may lose most of the particular charac- teristics of the parent plant, and may even cease to be productive. This could be a case of unstable hybrids. In the face of all these causes of variation, one can understand how the native himself gets lost.. In trying to recognize as great a number of different pandanus as possible, we took in turn, round the villages and babai pits and also into the bush, several men who were reputed to have a good knowledge of the different varieties. a50= They could with little hesitation, give names for their own trees but they were much less categorical about those of others. In the bush, their hesi- taney increased and they were often unable to assign any particular name to any particular tree. Further, when these "specialists" were brought to- getner, interminable arguments between them testified to their inability to agree on any specified term. They were not even in agreement on the differential value of a characteristic or a group of characteristics, which we for our part had difficulty in distinguishing, so subtle were they. Finally, we were able to retain only sixteen different names cor- responding to trees about which the natives we questioned did not hesitate. These were varieties recognized on the island of Tarawa. TARAWA LIST 1 - te antinakarewe 9 + te aratokotoko 2 - te anikomri ; 10 ~ te antibitia 3 - te aranaonimai 1l - te arateman 4 = te anabanaba 12 - te aratekura 5 - te arabaikiaro 13 = te aramboia 6 - te anikatouea 14 - te aramatakoi 7 - te arataitara 15 - te aramaiki 8 - te anibanakoi 16 - te aniwaentang In the southern islands, names as well as varieties may be different. On Nikunau we distinguished twenty-five names. Among them only te artinakarewe is to be found on the Tarawa list. In both atolls this variety is considered the best of all. NIKUNAU LIST 1 - te aramaru 14 + te irikeang 2 - te irikiri 15 - te irimaoruru 3 - te irimakiro 16 - te irikanoabuana 4 = te tina 17 - te irikauri 5 - te iroro 18 = te iribuangui 6 - te antinakarewe 19 - te n'tinatina 7 - te iritawatawa 20 - te anabanaba * 8 - te awaneari 21 - te irionotoa * 9 - te irinelaro 22 - te iriatabu * 10 - te aramaoia 23 - te irimangkoriki * 11 - te utongau 24 = te irikaiwete * le - te ararikitoa 25 - te tinanikarawa * 13 - te iribaoti a6 ie The names in the Nikunau list followed by an asterisk seem to have no meaning for the inhabitants of the northern part of the island. A fruite less variety was also mentioned to us, which has no usefulness other than as a source of a substitute for cigarette papers, but we were unable to check whether the name given, te kamoneara, is actually applied to a variety of tree or only to a type of leaves. When we study the various uses of the pandanus we shall see which are the most useful species and the advantages of each (p. 59). Ecological Survey: In our examination of the pandanus trees growing wild in the bush we tried to determine whether the sites they occupied corresponded to zonations influenced by existing environmental factors. These attempts never resulted in valid conclusions. The lagoon side, ocean side and interior of the islands have about an equal density of Pandanus, not for any one area con- sidered separately, but for a whole atoll. Noting that pandanus trees de- velop more fully in areas where tall vegetation is rather sparse, one might think that light plays a major role and that these pandanus grow so well in such areas because they are not impeded by the heavy shade of the coconut palms. Undoubtedly the light factor is in their favour, but other elements intervene to explain their abundance in certain places. Thus here and there in the center of islands, one observes areas which appear to be clearings in the thick undergrowth, where coconut palms are rare or absent, either because those destroyed by age or fire have never been replaced or because the environment has been too unfavourable. The pandanus, less demanding and more hardy, have developed and become the predominant element here. They had to compete only with slow growing species such as Guettarda, Scaevola, etc., which they rapidly outgrew. Even when fallen nuts or very young palms were present, the survival chances of the pandanus were greater. Their plentiful seeds germinated quickly, their early growth often being helped by the protection of the shrub vegetation, and during the first months of their existence a few rains were sufficient to ensure their rapid development. There are, moreover, other areas where the pandanus has definite ad- vantages over the coconut palm; for example, in places which by their nature or situation, or because of prolonged drought, are so arid that even the low vegetation suffers and often disappears. There again, the pandanus manages to survive. Such is the case also for marshy regions or for lagoon embayments where the water is not frequently renewed. These areas are regularly subject to intense drought, sometimes for very lengthy periods. Salinity, no longer compensated by rain, becomes so high that the coconut palm cannot resist it, although the pandanus tolerates it. Thus it is by no means rare to see dead coconut palms side by side with healthy pandanus, the latter having benefitted from the presence of the water without suffer- ing from the salinity. Apart from these special cases - such situations usually being restricted topographically - the pandanus is everywhere present, either isolated or in groups, in the middle of the most diverse environments (the mangrove excepted), or in pure stands which are often very dense (Plate Xa). 52 Finally, the human factor may intervene in favour of the extension of | pandanus or against it. Here, an owner will never clean his land, and pandanus will freely multiply on it (we frequently saw veritable "natural nurseries" of young pandanus in untended areas). There, ancther native will clear his property, removing these seedlings, with all the more satisfaction as he fully realizes the low value of pandanus trees raised from seed. Far from being limited to such special cases, man's influence on these small islands is at work generation after generation in a constant fashion. The parcelling out of properties and many other factors do not allow the formation of vegetation zones that would exist under normal conditions, nor have these been able to persist from before man's occupation of these islands. Cultural Techniques; Every pandanus planted by the Gilbert Islander is from a cutting, for reproduction from seed offers nothing but disadvantages, as it never transmits all the characteristics of the parent plant. The best cutting is obtained from a branch which already bears the beginnings of | adventitious roots. Generally a hole not exceeding 30 em in depth is made, | and the cutting is inserted without further care. Such a rudimentary tech- nique is regrettable, for many examples prove that pandanus always fares well after deep planting, with the addition of some fertilizer such as dead leaves, surface soil, or better still, compost. The more careful nativec often collect, in the bush, the black surface earth under the Guettarda and fill the hole with it. They even add various fertilizers as they do for the breadfruit tree. On the islands affected by the war, where quantities of scrap metal are found, the natives will often add a piece of iron in the hole. The useful- ness of this practice is debatable since pandanus seems to be just as pro- ductive without it and, moreover, never shows the slightest sign of chloro- sis. A planting technique to be noted is the frequent tamping of the soil around the just-planted cutting, considerably increasing its chances of suc- cess. In the case of the Gilbertese pandanus two things, however, remain rather difficult to explain. One is that this tamping is done several times, long after the plant has effectively taken root; the other is that the native insists that this tamping enables him to obtain both a very low tree (Plate Xb) and a more plentiful production. We have noted the accur- acy of this first point, but have no proof of the second. A further native assertion is that even if cuttings planted in the villages come from tall pandanus of the bush, trees grown from them remain small if tamping is car- ried out conscientiously. Growth: In spite of much cross-checking it is only with every possible reserva- tion that we quote some figures supplied by the natives. A period of less than a year (nine to ten months was even mentioned) from the time when a abit cutting is put in the ground and well packed is necessary to obtain the first fruit. A native on the atoll of Tarawa planted a cutting of the aratokotoko variety, a mile or so from his village in a large open space. He buried it in a hole about a foot deep which he filled with coconut husks, mesocarp, and other rotten vegetable waste to which he added rusty iron. A year later he passed the spot and was much astonished to see al- ready a large fruit (Plate Xc). Such astonishment 1. »* us think that pro- gress must usually be less rapid and that it would be reasonable to allow a year and a half rather than a year, with from time to time, exceptional cases of more rapid development. On several occasions natives told us that pandanus may produce fruit after only ten months if they have been planted in villages, but that, planted in the bush, they need more than a year. If this is accurate it can be explained both by the fact that the open situation of villages af- fords them more light, and also because all village trees are better tended. The period is reported to be relatively the same for all varie- ties. Hardiness: In spite of the probably very diverse origins of many "varieites" we may assume that they have all an equal degree of hardiness. One cannot but be struck by the resistance of the pandanus to salinity (Stone 1951 b, p. 18), by its adaptation to the most arid soils, and by its capacity for resisting extreme drought on areas subjected to intense sunlight and often to frequent showers of salt water spray. It must, however, be pointed out that in certain very exposed areas such as the corners of islets, and where there is a combination of many unfavourable conditions, even pandanus will succumb to the effects of pro- longed drought. Productivity and Longevity: According to the natives of the southern islands, fruit production is spaced quite regularly over the whole year, whereas Gilbertese of the cen- tral and northern islands state that there are clearly-defined periods of good, medium and poor production. The best period is July and August; production thereafter decreases steadily until Christmas, remaining very low until April. It finally rises again progressively from May to July. In the absence of statistics, it would be very risky to try and interpret these two viewpoints, in spite of our knowledge of differing rain condi- tions for the south, centre and north. ie The only facts which we ourselves have been able to establish relate to the very low rate of production of pandanus trees which had just suf- fered two years of extreme drought, and the plentiful production of these ~Shhe same trees after less than three months of regular and abaundant rain, with larger fruit than in the dry season. The natives agree with the assertion that the quality of the fruit is unvaried throughout the whole life of the tree, whatever the conditions of the moment. Since the uses of the pandanus are not confined to its fruit - its leaves and wood have a very important economic role -we must add that the behaviour of these trees in favourable rain conditions is rapidly marked by greater individual development and notable lengthening of the leaves. The natives of the southern islands, however, told us that the decrease in size of the leaves due to prolonged drovght was never such as to hamper the manufacture of the various handicraft products. The productivity of the pandanus lasts for many more years than that of the breadfruit tree, but no average figure was given us. During periods of intense drought it may become very low, but it never falls to zero, as in the case of the coconut palm. As for the longevity of the pandanus, it is commonly said that it has the normal duration of a man's life. Harvesting: As with the breadfruit tree the fruit, if too high to be picked by hand, is gathered with the help of a knife firmly attached to the end of a pole (te butika); the fruit is detached by cutting the stem. If the tree is very high, an adjoining tree is used to get near enough, or if the pandanus is isolated, it is climbed. There is never any waste. Rarely is a fruit lost, and then only because it fell unnoticed into the under- growth. The leaves are gathered preferably in the early morning because they have still a little of the flexibility which night has given them (marau), and are consequently less prickly. If gathered in the heat of the day it is impossible to avoid the cruel scratches frequently inflicted by their serrated edges. Uses: Fruit: Sabatier (1939, p. 34) tells us that the pandanus of the Gilbert Islands produces better fruit than any other in the Pacific Islands be- cause it grows and ripens slowly, and he adds that in Oceania there is no race which eats as much of it as the Gilbertese. In addition to form- ing a reserve food supply the pandanus fruit is consumed under different guises more or less daily throughout the entire fruiting season. The drupes, or more accurately their fleshy basal part, are eaten raw or are included in various dishes which are improved by the addition of grated coconut. Some recipes rarely vary, but others depend on per- sonal taste and available material. We shall indicate only the commonest: -55- Te tangauri: This is a preparation intended to be eaten quite soon. Composition: 2 parts of drupes (juicy section only), 1 part whole drupes and grated coconut (to taste). Preparation: The whole drupes and the juicy sections are cooked to- gether, then crushed, either with a wooden pestle, which is the ancestral method, or in a meat grinder, to obtain a purée. First, however, any parts of the drupes which are too hard are carefully removed. This purée (unlike te tuae) remains full of fibrous ele- ments. Grated coconut is added to taste or to achieve a certain consistency, and the whole is well mixed. This "fruit paste" is then eaten either in this state or later in the form of flat cakes obtained by drying the paste in the sun. This task, reserved for the women, consists of spreading a mat on the ground and arranging on it, side by side, the largest leaves of te uri (Guettarda). The paste is spread on these leaves with the tip of a knife to a thickness of about 4 cm. and is then covered with another layer of leaves to pro- tect it from flies (Plate Xd). This tangauri cannot be kept long because of its fresh coconut ccntent. It could be kept only by toasting the cakes, as is done for the preparation of kabubu. Te tuae: This is a preparation intended for keeping a long time. Composition: The juicy section of the drupes only. Preparation: Pandanus fruit is cooked in the native oven (1) by steaming (the method called te umum) or much more simply in a container such as an enamel bowl placed on a wood fire. After cooking, the edible part of each fruit is scraped. It is well mashed and the maximum quantity of fibre removed, and the resultant paste is spread as in the above recipe. When it has ceased to stick to the leaves it is cut into flat cakes which are exposed to the sun, first on a plank or iron sheeting, and then directly on the ground . (Plate XIa). Once desiccation is complete these cakes will keep for years. They form the reserve food supply with which the Gilbert Islander provides himself when he is travelling from island to island or going for a (Tieserintions of this oven and its installation may be found, with sup- porting illustrations, in several classical works such as those of Grimble. -56~ Te karababa: Composition: Preparation: Te kabubu: few days to the neighbouring wission's festivals. When he wants to use the cakes he soaks them first in coconut milk to soften them again. They are also eaten rolled up with grated coconut, then cut into thin slices and moistened with coconut milk (as we should do for French pancakes rolled with jam). Another way of using tuae consists of rolling it, cutting it into small pieces like noodles sprinkling it with kabubu (pandanus flour) and pouring kamaimai (toddy molasses) plentifully over the whole. This dish is then called te katii. The varieties of pandanus which are preferred for te tuae are (Tarawa names) te antinakarewe, te arabaikiaro, te arataitara, and te aratekura. This is intended to be eaten very shortly after prepara- tion. . Two parts of pandanus fruit; one part of grated coconut. The edible part of the drupes is cut, cooked as for te tuae end mashed without removing the fibres. When this | mass becomes creamy, grated coconut is mixed in and the paste spread in the sun in the same way as te tuae. Te | karababa may be eaten either when it reaches the con- / sistency of jam, or as dry cake, but generally it is ) made into kabubu. The cakes of karababa are toasted on both sides over a kind of grill made of flat heated stones. They are left to cool and then grounded in a large Tridacna — shell (te aubunga). ‘The coarse flour thus obtained is called kabubu. Nowadays this product is preserved in tins. Formerly it was simply rolled up and wrapped in pandanus leaves (te iria). It is said never to get mouldy. This is hard to believe for people who lik ourselves witnessed for months the almost daily renewal of mould on food, clothes, shoes - and the herbariun specimens which had constantly to be redried. Four years is claimed to be the normal keeping period of te kabubu and te tuae, if placed in tightly covered con- tainers, but many natives told us that they will keep as long as required. A last-minute item of information from L. Hudson confirms this. Kenneth Emory found in the Bishop Museum one of these iria which contained kabubu in a good state of preservation after fifty years. Kabubu may be eaten dry, as for instance by fishermen who remain for hours away from their home base. It is also put into a number of dishes, one of which - a particular favourite - is called te korokoro. i Te korokoro: Kamaimai (toddy molasses) is mixed with kabubu until it forms a kind of very thick cream which is eaten just as it is. Our Gilbertese interpreter, Te Baara, added that one needs a drink of water afterwards. In fact it is more usual to mix kabubu with water or toddy and drink it. All things considered, kabubu, although in itself rich in food value, is above all a base or "support" for other products, and it is these toddies and molasses which make it appetizing. Food Value: Actually, there exist very few analyses on which one can rely to es- timate the food value of pandanus. Authors are agreed on the considerable importance of the fruit and its derived foods as emergency food supplies, but they indicate only the high vitamin C content of the flour, te kabubu. Neither are there any data on the amounts consumed. The Gilbertese eat pandanus at their rate of production, and consequently consumption is high in times of abundant fructification, as with breadfruit. Leaves: Because of their multiple uses pandanus leaves have, in native life, an importance at least as great as that of the fruit. Roofing of houses: The minimum life of a roof (Plate XIb) is three years. Partitions and Walls: (Occasionally the adventitious roots, split down their length, are used for this purpose.) Mats: (a) Floor covering) ) The whole width of the leaf is used. (b) Sleeping mats ) (c) Dancing Young leaves are mostly used. Contrasts ; of colours are obtained by different treatment of leaves, by dyeing, or even simply by using leaves of different varieties. Headgear: (a) iWide-brimmed hats (fine weaving). (b) Conical fishermen's hats (coarse weaving). Sails: For canoes (pandanus leaves tend more and more to be replaced by canvas). Wrappings: For food supplies. -58- Caulking Material: For canoes (tends more and more to be replaced by European material). Baskets for babai: Whole leaves. Cigarette Paper: A useful substitute for cigarette paper is ob- tained by splitting the thickness of the leaf. Favourite varieties for mats and generally for all finer handicrafts are: White leaves only: Te antinakarewe White and yellow leaves: Te aratekura and te arabaikiaro The same varieties are preferred for cigarette paper. Further uses: The inedible part of the drupes is used as fuel. The natives perfume the coconut oil they use as hair oil with the spathes enveloping the in- florescence (te taba). The straight trunks provide framework pillars for houses (Plate XIb) and various supporting sections in the construction of maneaba (large communal house). The natives remove the bark from a piece of the adventitious root and pound the ligneous part. The juice is then extracted and mixed with four parts of water or of coconut water, and the concoction is used as a febri- fuge. Summary of most useful varieties: We have distinguished five different levels of quality among the 16 varieties clearly recognized by the natives of Tarawa. 1. Te antinakarewe: This is by far the best. A juicy fruit, fleshy and very sweet, it is the favourite for eating raw. Excellent for all culin- ary purposes; gives the best straw for handicrafts, white and pliable be- . cause drawn from long leaves; much valued for cigarette paper for its taste and its thinness. In the south islands this variety is also placed first. Karewe is the name for toddy. A linguist could tell us if there is any analogy with the taste (something sweet, smooth) or with the im- portance of this variety in the preparation of foods. 2. Te anikomri: Very useful as building material because the trunk is very slender, very tall and with few branches. The leaves are valued for handicrafts. Fruit is much prized for the preparation of te karababa and te tuae. 3. Te aranaonimai: Fleshy fruit is much prized as food, raw or cooked. -59- 4h. Te anabanaba, te arabaikiaro, te anikatouea, te anibanakoi, te arataitara. These five varieties are considered equally good for the preparation of te kabubu and te tuae. The fifth is more usually reserved for te tuae. 5. Te aratokotoko, te aratekura, te aramboia, te aramaiki, te aniwaentang, te antibitia, te arateman are the least valued varieties. Pests of Pandanus: The study of pandanus pests would require intensive work by a special- ist in the field. We have just seen the great importance of the leaves with their many uses, and it is they which are subject to considerable damage by a burrowing larva. These attacks are revealed by the appearance of a spot, 4 to 20 em. wide, whitish with brown or réddish edges, result- ing from the necrosis of the leaf, between the two epidermal layers of which the larva lives. On both sides of the leaf over the affected area the only thing left of the blade is a kind of parchment, extremely thin, nd in the form of a blister or swelling. The larva and adult insect still await identification by specialists. Another larva, the adult insect of which could not be obtained, is sometimes found associated with the first, or installed in an old cavity as a secondary commensal which need not be considered as a pest. It feeds on dried vegetable matter. This larva reminded us very much of that of Callirhipis. -60- CHAFTER 6 THE BREADFRUIT TREE The breadfruit (Artocarpus ) is a tree of the Moraceae family. Origi- nating in the Sunda Islands, it is nowadays widespread throughout the inter-tropical zone, particularly in the Pacific islands. It grows rapidly, up to a height of 20 m. The fruit is edible, both pulp and seeds. However the fruit of certain species or varieties, often regarded as the best, are seedless. The breadfruit is well adapted to atolls, and its distribution is cer- tainly directly related to the salinity of the ground water. If it is not too crowded by other trees, the breadfruit forms very low branches and de- velopes a spreading crown. It is planted in villages and along roadsides. The fruit season lasts from May to July, but may be prolonged wtil December. On Arno (Marshall Group) during the fruiting season, the Arto- carpus forms the basic food supply. The fruit of the mai, when very ripe, is sweet and has a fruity taste. The Breadfruit on the Gilbert Islands. In the life of the Gilbertese these trees (te mai) play an important role which in some regions of the archipelago may even rival that of pan- danus. They are, however, much less common. Whereas pandanus is one of the constituent elements of the "bush" flora and may at times form pure stands (Plate Xa), the breadfruit is found only in the villages or their immediate neighbourhood, and occasionally by roadsides, but never in the heart of the bush. In the villages the trees are automatically well spaced by the houses. The breadfruit tree is more widely cultivated and more carefully tended in the southern islands where, because of harder living conditions, its presence is more useful than in the central and northern groups. There can be no doubt that of all economically valuable trees Artocarpus is the one which has the least resistance to prolonged droughts. Sabatier writes: "In the south islands group, breadfruit trees survive with difficulty. They are practically exterminated every ten years." We can state as a fact that after two years of drought, many trees were dead or dying in the north and centre as well as the south. On Abai- ang the percentage of Artocarpus which had died in the course of the pre- ceding year or which were dying was estimated at approximately 60%. It was the oldest trees which were disappearing (trees which, we were told, were at least forty years old, a few even fifty). On Abemama few trees had died, but the tops and some secondary branches half way up had withered (Plate XIIIa). Similar observations were made on Tarawa. The islets at the northern extremity had suffered less than the rest of the atoll. On Aroraeé some areas were badly affected, others hardly at all. On Onotoa there were many dead trees of widely-differing ages (there was no re- planting on the return of the rainy season, for want of seedlings. The ~6]l= main reason for this lack is that Artocarpus seeds lose their germinative properties fairly rapidly ).On Nikunau there was a nwiber of dead trees, while others had only dried tops. On Beru, there were few dead trees as compared with the other islands, except in the southern part of the atoll where trees of all ages died, the proportion being approximately 60%. In spite of its lack of resistance to drought and to variations in the salt content of ground water, the breadfruit has remarkable regenerative and recuperative powers (as shown by the breadfruit trees of Butaritari which were damaged in the war). Often the trees make a new start by throw- ing out a shoot at the base or half-way up the trunk, but these “resurrec- tions" depend above all on the environment of the root system. Thus there is great contrast between the excellent survival of some trees and the dying state of their neighbours. Here as with coconut palms water is the important factor. Because they happen to be near a fresh-water well, individual trees, having survived intense prolonged droughts and reached an imposing size, live to a great age and maintain a remarkable rate of productivity. At the old site of the village of Bita (Tarawa Atoll), about half a mile from the present site selected some ten years ago, only two trees sur- vive out of the hundred or so which were there before the place was abandoned for lack of fresh water. Finally, to the topographic and edaphic conditions influencing survival must be added the individual quality of the trees and the attention they have received. Species or varieties: As a result of our own observations and enquiries among the natives, it is possible to distinguish five different Artocarpus and the botanists who are examining the specimens will indicate whether they are species or simply varieties. Te mai: (Plate XIc). This name is the general vernacular term for any breadfruit tree in the Gilbert Islands (also in other areas of the Pacific) and in addition designates an Artocarpus of which the distinctive charac- teristics are: Leaves: Pinnatilobed. Fruit: Ovoid with a more or less pronounced median transversal "narrowing". Numerous seeds. Of all the Artocarpus te mai is the most appreciated by the natives. It is also the most prolific. Te maitarika: This name may be broken down into mai (breadfruit tree) and tarika (brackish water). This is the hardiest Artocarpus and it is able to exist under more difficult general conditions than the others. Above all, it is more resistant to brackish water. Distinctive characteristics: (Plate XIIa, b). -62- Leaves: Trilobate extremity - median lobe larger than two lateral lobes. Fruit: Pyriform, dented rather than "narrowing". Fewer seeds than in te mai. Te maitarika is the least tall variety, Te _keang ni makin: (Plate XId). This name indicates the striking resem- blancs between the leaves of this Artocarpus and those of te keang, a fern (Polypodium scolopendria Burm. f.), which is fairly widespread in the Gil- bert islands, and the supposed origin of this breadfruit tree (the Islands of Makin). Distinctive characteristics: (Plate XIIa, b) Leaves: Pinnatifid. Fruit: Resembles that of the te mai rather than that of te maitarika, but it is more commonly ovoid and contains fewer seeds than te mai. Te Keang ni makin bears later than the others. It is the least common variety. Te bukiraro: The characteristic feature of this Artocarpus is the absence of seeds. It is, therefore, always propagated from cuttings. It is usual- ly taller than the others. This variety is very productive and is the favourite of Europeans. Distinctive characteristics: (Plate XIIc). Leaves: Pinnatilobed ~ much like that of te mai, but wider. Fruit: Almost spherical, neither narrowing nor dented. Te motini wae: This Artocarpus is also seedless and, like the previous one, is reproduced by cuttings. The name contains the word te wae, which is also used for coconuts without kernel (te ni wae). This variety is less common than the others. We saw it only on Betio, growing from a cutting which had been brought from the island of Butaritari two and a half years previously. It was a tree about fifteen feet high and had no fruit. Cultivation: The term cultivation may appear a little exaggerated. However, the breadfruit tree does receive a certain amount of attention, both at the time of planting and throughout its first years of growth. A hole is dug, usually 30 cm by 30 cm and 20 to 30 cm deep {on Abaiang we saw holes 1.5 m deep). For several weeks the hole is filled with all possible waste -63- matter of which the most abundant is vegetable matter, dead leaves of coconut palms and of Artocarpus, and with all the rubbish swept up--daily. Leaves of te mao (Scaevola sericea) and of te non (Morinda citrifolia) are, however, preferred. A little earth is put on top and the seed is sow. If it does not germinate another is sown later. The pit is seldom watered but the young sapling is nearly always pro- tected by being completely enclosed, in the beginning, with leaves or a mat (Plate XIId). Later the roof of the shelter is removed and only a circular fence is left, against very small children, piglets and fowls. Naturally the accumulation of vegetable waste entirely filling the hole subsides progressively as it gradually rots. The native compensates for this shrinkage by adding dead leaves and other matter which serves at the same time as fertilizer. Often, as a protection against animals, as well as to keep the compost in place, a’ wooden frame made of four coconut logs (Plate XIIIa) or a border of small coral blocks is built about a meter from the foot of the tree. In some villages of Abaiang we noted that the refuse from the pig enclosures was collected and put inside this fence. Finally, in many cases, the hole was topped with a layer of the very fine black sur- face soil which is found in the bush under te uri (Guettarda speciosa). In the Gilbert Islands the breadfruit are better respected than other trees, we mean that their trunks do not show the multiple gashes which the native is in the habit of making on coconut palms to carve his name. Under good conditions these trees sometimes reach considerable dimensions. One we saw was almost 18 meters tall and was much divided near the base, each branch more than 1.5 m in circumference, 1.5 m above ground. Some trees may even exceed 20 m in height with trunks over 2 m in circumference, and are said to be more than sixty years old, but such are the exception. We were told that trees over fifty years of age suffer more from drought than younger ones. We question this however, since for these trees to achieve such an age they must have enjoyed exceptional conditions, which should have helped them resist drought. In our opinion they must have reached the end of their life-span, and drought merely hastens their death. Under less favourable conditions breadfruit trees from eight to ten years of age may be found, 8 to 10 m in height and 1 m in circumference at 1 m above ground. The number of Artocarpus varies from one region to another, and from one village to the next. It is generally dependent on the number of in- habitants although on certain southern islands we noted a higher mean number of trees than elsewhere for the same population. In a village on Tarawa Atoll (see table X) for a population of 115 persons (22 families) we counted 93 breadfruit trees; only one family did not own any. In the Gilbert Islands these trees are grown only in the village. They always belong to the person who planted them and even if he moves elsewhere he always has prior right to the fruit. | ake Production: Obtaining production figures per tree or per village, is not possible. No one has ever thought of weighing, throughout a normal and dry year, the fruit of one or several trees. The only information of any interest is the fact that in a dry period they are sparse and reduced in size. In July, when the rains had lasted more than three and a half months, five fruits picked from an average tree (te mai) from Tarawa gave us the following Meee reR ib teers sc it 2b. /9 eas). 1 1b. a ogc.) 1b OG oge.> L Ib. loz. It is towards the end of the year that the trees bear their maximum of fruit. According to our native informants, the production of the bread- fruit tree lasts nine months of the year, with a slack season of three mouths in February, March and April. This pause which, according to the Gilbertese expression, corresponds to the tree's resting period, may ob- viously vary with the rainfall. In the Society Islands, Papy (1951) re= ports a full harvesting season from November to April and another fructification in July and August. The fruit is detached from the trees with a long pole (te butika) on the end of which a knife blade is fixed. If the tree is very tall the gatherer climbs it, cuts the stalk and the fruit falls, without suffering much damage. It is against customs to have women gather breadfruit. Consumption: The delicate flavour of this fruit makes it a favourite food of the Gilbertese. In times of abundant production, when too many fruits ripen at once, some will be made into preserves so as not to be wasted. As an item in the daily food supply, the fruit is eaten raw or cooked, after the skin has been scraped with a knife or shell. It is cooked in a little water and steamed or baked in the standard native stove covered with mats, or alternatively as we cook potatoes "in their jackets". In the first case, the cooking water is never thrown away; it is much prized as soup. Preparation for keeping: The preserved product is called te kabuibui ni mai or te tuae n-te mai. The fruit is cooked in its skin or first scraped with a te koikoi shell. The slight cooking is done in a large enamelled basin, which is less work than setting up a stove in the old manner. The edible parts must then be crushed. A giant clam shell, te aubunga, and a wooden pestle of te ngea (Pemphis acidula) were formerly used; we saw them still used by very old women on Onotoa. Nowadays, the Gilbertese who have been working on Ocean Is- land or Nauru often bring back meat grinders. The use of these tends to be- come more general. Obviously there is not one for each family, but they are San ie like many other things, by the traditional bubuti (see pp. 16-17). -65- Insect Pests: The Gilbert Island Artocarpus seem free of insect pests. We never noted any attacks, even slight, on the leaves, trunk or branches. How- ever, from an identification list just received from the Bishop Museum we note a fruit fly, Dacus (Strumeta) frauenfeldi Schiner. Pharmacopeia: (a) Ear Complaints: To soothe certain painful affections of the inner ear, four or five young Artocarpus leaves are crushed, the juice extracted and poured into the ear. Alternatively a young leaf is warmed and stuffed into the ear duct. (>) Conjunctivitis: Artocarpus buds are chewed and spat on to a piece of cloth which is squeezed above the eye and allowed to drip into se Other uses: The wood of Artocarpus is used particularly for outriggers of canoes and for fishing floats. The leaves serve as fertilizer for the babai as well as for breadfruit itself. They are also very useful for wrapping many native foods for transport. In preparing kabuibui, the hands always become sticky and stained black by the latex which exudes from the ripe fruit. Soap and other European cleansing aids do not easily remove these stains. The Gilbertese places a small piece of copra in his mouth, chews it, spits the result into his hands, then rubs and rinses them with water. Every trace of stain and stickiness disappears. Food value: Authors agree that the breadfruit has high food value. It is esti- mated that it contains about the same quantity of edible carbohydrates as taro or sweet potato. When it has reached full maturity it gives, like the banana (Buchanan 1947), a negative reaction to the iodine test, show-- ing that all starch has been converted to sugar. The cooked fruit is a good source of vitamin C. =66- TABLE X Census of a village of Tarawa comprising 23 households. Se a ee ee ee ee Family, Identi- fied bys sulibets eo 3s 4 5 OF fe «0 - 5s 16 17 18 19-20 2 22 23 Total Number of persons (Sy Bp ee a ee ee eS Ce a ee No. breadfruit trees 2 eee pe Oa Oe 8 So Be ee SS 93 No. of Papaya : trees See ee ee ee ee OS 3 No. of pigs dE ipa SB SE Sa es SS ea ae 2 ee ee a ee ee es eee he No. of fowls Cee Oe ee De eG GS 205 No. of babai : pits eee ee Oe ee ee eS -) S ont No. of parcels . of land eee re ee a Be ee se 5S 108 TABLE XI Jumber of Cyrtosperma of 3 age sroups for 27 families in a Tarawa Atoll village. Families Adults Children Pits Cyrtosperma growing for 6 to 13 mos: 2 to yrs. 6. to 12 yrs: 1 2 nh 2 200* = - i 2 2 a hs 9 2 2 5) 3 3 oe 6 8 1 h = 3 117* 61 62 2 5 “ 1 al 22 138 2 hh zs h 252% 14.3 x 3 6 2 126 5h 8 i. 3 1 Ms 107 113 al i 2 = a ho6* 6 = ik 3 - 2 95 Lh 12 2 5 2 2 167 56 20 i h a 3 130 h ~ ab 3 2 3 9h 28 1h z 3 al i 53 16 7 z 5 a 2 1h6 23 10 i a 2 al 211* an bi 1 hi 1 3 1oh* 30 138 il 2 2 2 65 30 12 l 2 < a) 66 12 he 1 3 2 a LO 18 10 a 2 a 2 88 1) ) 27 if 29 hh 2681 730 331 * 80% are represented by the variety Te Katutu. CHAPTER 7 THE BABAI Te babai is a tuberous plant of the Araceae family which grows in stagnant water; the aerial parts reach 4 height of 3.5 m (Plates Life XIIIb, c). The leaves are extremely wide, with more or less thoray petioles. The edible tubers are often wider than they are long. Food value should approximate that of taro (Colocasia esculenta). Te babal is found in the Gilbert and Bllice Islands, the Caroline Islands, the Marshall Group, and in French Oceania. There are also small areas of babai on Fanning Island (Line Is.), Gardner Island (Phoenix Is.), Hull Island (Phoenix Is.), and larze areas of wild babai and taro at Washington Island (Line Is.). The assignment of a correct identification to the babai seems to have been, for a long time, the subject of great confusion. It has been fre- quently mistaken for: Colocasia esculenta (taro); Caladium cordifolium; Alocasia macrorrhiza or Alocasia indica. In 1927 Christophersen gave, ac- cording to ane the correct identification: ECyrtosperma chamissonis (Schott) Merrill (1), In the Gilberts this plant occupies an important place in native life. It is in fact the only crop actually cultivated by the Gilbertese. They grow it with the utmost care, but it should be noted at this point that the babai is not, as is generally believed, a staple food, at least nowa- days. There is no doubt that, in several islands of the Group, the babai was much more extensively grown in the old days than it is now, as evi- denced by the number of abandoned pits. In atolls such as Abemama the ground was literally riddled with pits which have been gradually abandoned and partly filled up by the sliding of the embankment slopes. The babai is sow almost a luxury. After living in close contact with the natives we came to the conclusion that, for the Gilbertese, the value of the babai seens greater from the psychological point of view than as a foodstuff. The growing of babai is, indeed, surrounded with a number of traditions in which pride seems to be predominant. For instance, much care is still lavished on specimens of which the only remaining value is that of ornament and the enormous size of their leaves. A native will display the sreat- est pride in a babai ten years old (or even older) which has reached an extraordinary size, although the tuber, if it still exists, has usually grown tog haird and bitter. Moreover, pulling up a babai in order to offer the tuber to-a distinguished guest is considered the greatest honour that can be paid to him. In more general terms, it will be easily understood (1) Our friend Professor BE. D. Merrill recently sent us complete tidi- cations which enabled us fully to clarify this identification. LSS that a plant so difficult to plant, which srows so slowly and requires so much tending, is a food that cannot be used indiscriminatel,. The size of the babai pits varies considerably with topographic and edaphic conditions. They may range from ten to several hundred square meters. Several families or several individuals often share a vit (see Table XI). The shares are quite unequal; one person may own ten plants or more, while another has only two. The areas belonging to each and in- volving ownership of babai plants are said to derive from ancient land ricshts. It is generally considered that a fully-grown babai requires one square meter of living space. Most Gilbertese pits are very ancient; we have no recollection of seeing one of recent origin or being prepared. The old oits are still used, and occasionally a pit which has long been abondoned is brought into use again. A fact should be mentioned which at first glance seems parado::ical. Some islands of the northern group, in which the food supply is best, have remarkable pits and a proportionate abundance of babai. On the other hand, some of the southern islands where rain is extremely in-. frequent have very few and very poor pits. This is a matter for regret, since a greater production of tubers would be a valuable offset against the food shortages which sometimes affect these islands. But it should be recognized that soil conditions are also against them. We shall see later that many pits are often given up because the water has become brackish. In other regions, pits were abandoned a long time ago, sometimes over very large areas. This was often the consequence of a demographic disturbance which resulted in olden times from local native politics {this is the case in Abemama, Aranuka and possibly Kuria). Brief comments on some of the atolls visited: ' Marakei: This is the island where the babai seems to be most extensively grown. There is a great number of pits. Most of the plants are over 3 m tall and are well and consistently cared for. Abemama and Aranuka: There are too few pits for the number of inhabitants, end most of them are not well kept. Abemama was, in the past, practically riddled with pits, but today coconut palms occupy the depressions which are all that remains of them; these are of course among the healthiest palms on the island. Most of these pits have been out of use for a long time, possibly since the end of Binoka'ss reign. In those days the babai may have had a greater ritualistic significance. There are also pits abandoned when water became brackish, as on Bike Island, now uninhabited. Tarawa: The pits are generally very fine and the babai flourishing; for instance, those growing in the villages of Eita and Bikenibeu are almost as remarkable as on Marakei. \ Arorae: The pits are remarkable, mostly because of the great amount of work required for their excavation. The rocky platform is often of exceptioaal thickness (Plate Id) and very compact. An enormous volume of earth some- times has to be dug out and the sides of the pits have to be laboriously propped up. A striking feature of this island is that the material dug from the pits through centuries has formed a mound on each side. Nikunau: In this island the pits are found mostly in the north, the pit water being said to become brackish in the southern part. The result is that the inhabitants of the south ask the northern people, as a reat favour, for their permission to dig a pit in one of their plots. The rather miserable condition of most of the babai grown on this island gives a painful impression, In most cases there are even no baskets around the plants. Maintenance is so often neglected that one wonders whether such carelessness reflects discouragement on the part of the native in the face of such a poor growth. Most answers to our questions led us to think that little interest was bestowed on these pits and their babais. Cultivation: Cyrtosperma is cultivated in pits because it grows successfully only when living in fresh-water marshy land. Owing to the very special ground conditions in the Gilberts the natives have to dig down to the rocky plat- form and through it in order to reach the ground water lens. This work often represents a considerable effort, since the platform is sometimes at a great depth. As a result, the excavated material occupies much space and is sometimes as high as 1.5 meters. The conglomerate layer is often so hard that it takes a crowbar to break it. The next step is the strengthen- ing of the pit walls wherever there is a risk of their collapsing, using blocks of muddy earth from the last excavation. Lastly, the bottom of the pit is levelled and the young babai shoots planted in it. As they grow, a mixture made of humus gathered under certain trees in the bush and of selected leaves is spread around them. A basket of pandanus leaves and stakes (Plate XIIIc) keeps this debris in place and also supports the plant. As more humus is added the increasing weight anchors the whole more firmly to the bottom. Thus this is really like growing plants in pots and - at least during the first stages - the plants are floating. If the roots, instead of spreading within the supporting mass, grew down into the deep mud, the plant would die. It grows only because it is rooted in an artificial and aerated mediun. When a pit abandoned in the distant past is restored to use, it has to be excavated and cleared of the soil which has inevitably slid into it, and the bottom must be leveled again. Instead of digging the whole of the area the natives start by digging holes of the required size for each babai plant, widening them as the plants grow, until, the holes having merged, the vhole pit will have the same level. The work has thus become much less toilsome, since it has now been spread over many years. The native always chooses for the site of his babai pit a location where fresh water is available. Contrary to the views of some authors, -09- the water of babai pits is not brackish. Indeed, it is often fresher than that of wells used for drinking or cooking. During prolonged droughts, if water becomes brackish the babai plants are said to wilt rapidly. Brackish water is definitely noxious whatever the age of a plant. Long immersion due to rain-water may be fatal too, but only to young plants. Fertilizers: Leaves from various plants and other fertilizing elements are brought to the babai pit. Each owner has his secret technique for planting aad tending the plant which he keeps from everyone outside the family circle, Imowledse of it is transmitted only from father to son. The secret is in the technique rather than the nature of the fertilizing material, this being obviously known to everybody since the choice is rather limited. Of these techniques very few were disclosed to us, but through scraps of in- formation we were given to understand that they sometimes involve tamping with the feet of the soil around the young babai, or selection of par- ticular seasois to use certain fertilizers, or sometimes the proportions of the various components used in the compost. The leaves mainly used, are, in- order of importance, those of the following plants: Te kaura: Sida fallax L. -- te uri: Guettarda speciosa L. -- te ren: Messerschnidia argentea -- te mai: Artocarpus sp. -- te wao: Boerhavia diffusa L. -- te kaura ai banaba: Wedelia biflora D. C. (this used mainly in Abaiang) -- te kanava: Cordia subcordata Lam. Less fre- quently used are: te kiaou: Triumfetta procumbens Forst. and te rao: Hibiscus tiliaceus L. (called te Kiaiai in Beru). The latter is rather rare in the Gilberts. To the leaves of these various plants is added a quantity of vegetable refuse, more particularly old pandanus leaves, pieces of rotten coconut trunks, etc. A special mention should be made of te kaura (Sida fallax). The leaves of this very common Malvaceae are very seldom brought green to the babai; when they are, direct contact with the plant is carefully avoided. The native knows that the fermentation of these leaves would pro- duce heat which would either kill the babai or be very harmful. If time is short - or if it has been impossible for any other reason to dry the leaves first - the native will take care never to cover them with other material when placing them around the babai that they may dry more quickly and remain well aired. Generally they are dried in the sun on mats in front of the owner's hut. They are then stored in a basket to be taken later on to the pit, mixed or not with other fertilizing material (Plate XIIId). To fertilize the young babai plants, well-dried leaves of te kaura are often mized with surface soil, preferably gathered under Guettarda trees, to which well-ground pumice is sometimes added. This preparation, on which moulds tend to develop, is stirred from time to time in the basket in which it has been mixed; we were told that it would thus be stirred, turned over and perfected for two months and longer before being applied to the babai. =7O= Once the young babai is set the native no longer works at the bottom of the pit if it is too flooded by recent rains. Stirring the water and treading too near the plants when the root systems are still weak, might tear them from their support. Later on, when the babai is strongly rooted in, the native will go and tend it even if there is much water in the pit. The tending will consist in inspecting the gabions, replacing a loose stake or pandanus leaf, cleaning the pit of any extraneous material such as fallen coconut fronds, and weeding out certain invading plants such as te mam (Jussiaea angustifolia Lam.); te maunei (Eleocharis geniculata (L.) R. & S. and te ritanin (Cyperus javanicus Houtt.) But the most important part is the "feeding" of the babai. The tubers are said to be much more tasty if the plant has been abundantly and regularly fed. A new batch of fertilizer every three months is therefore considered necessary. Propagation: Several methods are used, depending on the varieties, almost always vegetative; the seeds of varieties with fertile seeds are seldom used. 1 (a). The method most often used, and which is obviously the simplest, consists in taking suckers from the base of the mother plant and setting them according to the procedure described earlier. 1 (b). Under special conditions, when suckers are lacking or when the native is dealing with varieties such as te ikauraura and te ikaraoi which produce few shoots or only after three or four years, the artificial pro- liferation of shoots is stimulated: Te baku, the top part of the rhizome with the bases of the petioles and terminal bud, is cut from a babai uprooted for eating and is first planted in a part of the pit where it cannot be reached by water. If it were planted directly at the bottom of the pit in a water-logged environment it might rot before growing roots. It will be left in this temporary site until the first two leaves have unfolded. The natives know that at this stase the root system has already developed. They will then cut the terminal shoot at a given height, which stimulates the artificial proliferation of suckers. When these reach a normal size they are broken off, just as spontaneous shoots would be, and planted in the bottom of the pit. 2. Another method follows much the same lines as the above but instead of cutting the terminal shoot, the whole plant is uprooted and set permanently at the bottom of the pit. By this method, a rhizome is obtained in half the time required with spontaneous suckers. Insect Pests: Two pests should be mentioned which do not yet cause very serious damage, but are none the less a threat. One of these lives underground. It is a scarabaeid beetle, Pentodon sp. or Papuana sp. The larvae bore galleries in the older tubers. fie Fortunately this beetle is not commonly found in the babai; it is more frequent in the old banana plants (also grown in pits), where it bores galleries just under the base of the shoot; one wonders whether it has not been introduced with banana planting stock. It might have gone from there to the Cyrtosperma. The Tarawa natives state that this beetle is also found at the foot of coconut palms. (see p.28 ) The other pest lives above ground. It is a lepidopteron of the family Phalaenidae: Prodenia litura Fabricius. The caterpillar sometimes causes serious damage to the babai leaves, mostly on young plants. + may eat as far as the central part. This insect is unfortunately very widely distributed throughout the archipelago, as in many other areas. The fact that the natives seem to show no concern about this pest and the damage it causes does not exclude the necessity of undertaking as active a con- trol of it as possible, particularly as it also feeds upon a large number of other useful plants. Varieties: As in the case of the pandanus, it would be easy to enumerate many babai names, perhaps representing more or less typical variations or merely different mames which vary from one island to another. Our efforts were therefore primarily directed towards identifying well differentiated varieties. The criteria used by Tarawa natives to determine the different varieties were checked and are submitted as a practical key for identifi- cation (Table XII). Remarks on varieties: 1. Te anrairaki: also called te babue; medium height (not over 2u tall); not very common. Quality of tuber, good. Said to be eaten after three or four years. 2. Te natutububua: medium height; not very hardy. The fully grow plant is said-to produce no new leaves, but a considerable number of shcots (bubua - about one hundred). The tuber is vider than it is long (pail-shaped). Quality of tuber, excellent. Said to be eaten from the third year onwards. 3. Te kaikui: Very tall, reaching 3 meters. Tuber may reach a length of 00-70 cm. Quality of tuber, excellent. Said to be eaten from the fourth year onward. kh, Te ikaraoi: Extremely tall (sometimes over 3m within four or five years). it is the finest of all vpabai. It requires much more tending than the others, particularly when young. Few or no shoots (propagated by methods 1 (b) and 2). Said to have the largest tuber and to be eaten around the fourth year. Quality, very good. 5. Te oineke: Height sometimes exceeds 3 m within four or five years. sve I / Under surface of leaf desquamate, epidermis bearing lobiform appendices II / Leaf undulate and upper extremity of \ / petiole distinctly bent in "swan's pe neck" Te natutububua IIt/ Leaf normal A/ Petiole green a/ Petiole spiny lo/ Number of leaves normal ( +5) 1/ Young leaves \ sometimes spatulate Te kaikui RY 2/ Young leaves s Le always normal Te ikaraoi 20/ Number of leaves double ( +10) Te oineke b/ Petiole not spiny lo/ Seeds fertile Te ibuota 20/ Seeds unfertile Te katutu B/ Petiole yellowish orange Te ikauraura C/ Petiole yellowish orange only at the base Te tukuna Table XII. Varieties of babai in Bikenibeu, Tarawa Atoll. PAHs) ‘Naquasig fo sarijmu ayy OF Suip4oo>0 / WEE fo daqgnz ey} jo yybua| Ul Y4Q*Aoug Lv 813 WID~ 2104 z/\ yl3NNH 3k | a W4Ap- 3404 yNUNBINVLUWYN aL wip /¢ »VUNGINVIVWUNDLNYLA \ WUD VWONY Jt yy) VY 2u0g4SdaIq OF n VASNOT JLy Numerous leaves; it is by the number of leaves that natives distinguish } it from te ikaraoi, Reproduction by shoots or by seeds. Tuber is longer than it is wide. Said to be already good for consumption after six to nine months. It is the most precocious of all and grows very large in the shortest time. 6. Te ibuota: Plant erect, mediun height, not very common. Fertile seeds. Flower said to appear only after three or four years. Said to be edible after two or three years. Quality, good (a little poorer than te oineke). 7. Te katutu (pronounced kasusu): Mediun height (1-21 ). The natives say "man height". Very hardy, requires less tending than any other. Flower appears around the second year if the plant has been well tended. Con- tinuous abundance of suckers requiring thinning. Some are uprooted and replanted elsewhere. The tubers are eaten as early as after 9 to 12 months when their weight ranges from 0.5 to 1 kilo. These figures vary with the fertilizers applied at the time of planting. The tuber grows hard very quickly after flowering. Some pits contain up to 90% of katutu, which ' means that harvesting can be cariied on almost all year round. This I variety is considered as too common to be used on special occasions a (visits, weddings, Christmas, Easter, etc.); its quality and size are | below the proper standard for gifts. Quality, good. If the leaf stalk | is slightly yellow, sub-variety uraura. If the leaf stalk is quite green, | sub-variety roro. 8. Te ikauraura: Very tall, sometimes over 3 m within four or five years, erect, more hardy than te ikaraoi. May flower after two years only if the plant is very well tended. Tuber does not grow as large as that of te ikaraoi. Quality, good. 9. Te tukuna (pronounced sukuna): Medium height. Very similar to te oineke. Quality, good. The species of babai observed most frequently are: Te katutu, te ikaraoi, te oineke and te ikauraura. Of the following varieties found on Nikunau (south), the four under- lined varieties are similarly called on Tarawa Atoll: Te inawaro, te kairoro, te kaiura, te unikai, te tebon, te taninganibuki, te babue, te ikaraoi, te kaiku, te katutu. Of those found on Marakei (north) three identical names (underlined) are found: Te moaine, te atinimainuku, te iokanai, te ikauraura, te ikaraoi, te katutu. A last variety was mentioned to us several times, but we were never able to see it: te n-touman, the tuber of which cannot be eaten until it is over four years old. It is said to be the slowest-maturing variety, to re- semble te ikaraoi, with shorter leaf stalks, and to be similar in taste. Stages of Growth: We were given by the Gilbertese descriptions of five different sizes of babai, which they measure with their arms. These observations were adapted to a growth curve (fig. 17). Te kunei corresponds to a plant approximately nine months old. At this stage the tuber would be very tender. Te katutu is the variety most commonly eaten at the kunei stage. Some varieties are highly prized at the te namatanibura stage, which corresponds to a growth of approximately three years; other varieties are much less relished at that stage, ‘the tuber sometimes being too bitter. The stage te namatanibura corresponds to a fully mature babai. The natives state that from this stage on the quality of the tuber decreases. The te anga stage refers to a plant approximately seven years old. Customs and traditions require that for certain rituals a babai of this size be selected. This is also the case for te bonaua, which may be estimated at.ten years old or more. The tuber, which has growa very hard and of considerable size, is a highly-regarded present from a young man's family to the family of the girl he is to marry. Often a babai of this age has reached such spectacular development that it is kept in the pit merely for pride or, according to some people's beliefs, be- cause it will stimulate the growth of the young plants growing near it. Food value and importance as food: Authors recognize the high starch content of Cyrtosperma tubers but they insist that they should be eaten only when perfectly cooked (boiled or roasted). All taro are generally considered as being poor in vitamin B and proteins. To assess accurately the consumption of a given number of versons could only be done by continuous checking on the spot, which woulda be difficult. As the babai is the only crop requiring real work from the native he does not generally eat it more often than three or four times @ month and increases this consumption only in those periods of the year when pandanus and breadfruit are scarce. Methods of Use: Foods: Te buatoro: For this dish raw babai is grated. The paste thus made is wrapped in leaves of te mai (Artocarpus) or even te babai, divided into portions, each portion weighin; approximately 1 1b. These bundles, tied with strips of fresh coconut leaflets, are immersed in boiling water. The cooking pot is placed at the edge of the fire and left to simmer for approximately 45 minutes; the bundles are then taken out of the water, opened and the preparation can be consumed as is, or mixed with kamaimai (toddy molasses). ee Te bekei: Te buatoro is crushed in a pan, then covered with coconut cream which has already been prepared and heated. The mixture is well stirred and may be consumed at this stage or when cooled. Gourmets will pour kamaimai over cold bekei. Te tangana: The babai is peeled, cooked, then crushed. In days of old it was crushed by means of a wooden or coral pestle in a giant clam shell. Today a meat grinder is more often used. Kamaimai and crated coconut are added to the crushed babai. Imported flour may also be added to make kneading easier, but this is not essential. The mixture obtained is wrapped in babai or breadfruit leaves and set in the native oven where it remains overnight if it has been put in late in the evening. This preparation may be kept for three or four days. If it is to be kept longer, it is cooked again every two or three days to avoid fermentation. Pandanus is sometimes added to this food. Te kabuibui ni babai: This preparation is the result of a process de- signed to insure longer preservation. The pieces of babai to be preserved are first parboiled, then crushed or, better, put through a meat grinder. The ground product is dehydrated by sun drying, and this kabuibui may be kept a whole year or even longer. The same technique is also applied to breaéfruit, then it is called te kebuibui ni mai. Fruits of te bero (Ficus tinctoria Forst.) are prepared in a similar way and the dehydrated preserve is called te kabuibui ni bero. In the southern islands these recipes of kabuibui appear to be more commonly used than elsewhere due to the more serious effects of intense and prolonged droughts. Pharmacopoeia: Enquiries into native pharmacopoeia too often yield little or no infor- mation, unless the investigator lives for a long time among the natives. The babai may be used for more than one treatment, but the only one dis- closed to us is a formula for curing skin diseases (though no one could tell us which or show us a case.) The tuber is cut in slices, which are laid in the sun. Yellow mould is said to develop, which is carefully scraped off and sprinkled over the sore. The gathering of this mould is the prerogative of only a few specialists on each island, and is considered a very delicate task. We could not obtain a specimen of this mould and did not succeed in isolating it. Such a variety of moulds developed on the slices of Cyrtosperma which we placed in various conditions that the natives themselves could not tell us which was the one generally used. -(5- CHAPTER 8 OTHER PLANTS USED BY GILBERTESE There are many other plants of economic significance used by the Gil- bertese for a variety of purposes. These are listed below alphabetically under 9 headings. These lists include some plants very scarce in the Gil- bert group but which will be, perhaps, useful later to the natives. Indi- vidual notes on these plants and their uses will be found in Chapter 9 Part x ae ; 1. Food plants Acalypha amentacea Passiflora foetida var. hispida Boerhavia diffusa Pemphis acidula Carica papaya Physalis angulata Cassytha filiformis Portulaca lutea Citrus aurantifolia Portulaca samoensis Colocasia esculenta Saccharum officinarum Cucurbita pepo Sesuviun portulacastrun Ficus carica Tacca leontopetaloides Ficus tinctoria Terminalia catappa Ipomoea batatas Musa paradisiaca subsp. sapientum 2. Plants used as building material for houses, canoes, fishing rods, etc. Bruguiera conjugata Lumnitzera littorea Calophyllum inophyllum Messerschmidia argentea Casuarina equisetifolia Morinda citrifolia Clerodendrum inerme Pemphis acidula Cordia subcordata Pisonia grandis Dodonaea viscosa Premna obtusifolia Ficus tinctoria Pseuderanthemum carruthersii Guettarda speciosa Rhizophora mucronata Hernandia sonora Scaevola sericea Hibiscus tiliaceus Terminalia catappa 3. Drug, medicinal and poisonous plants Barringtonia asiatica Fimbristylis cymosa Boerhavia diffusa Messerschmidia argentea Bryophyllum pinnatum Nicotiana tabacum Calophyllum inophyllum Pemphis acidula Clerodendrum inerme Ricinus communis Cordia subcordata Seaevola sericea Crinum esiaticum Triumfetta procumbens Euphorbia chamissonis 4, Plants used for ornamental purposes (mainly for making leis) Abutilon indicum Hibiscus tiliaceus Acacia farnesiana Lantana calilara Acalypha amentacea Lumnitzera littorea Acalypha wilkesiana Mirabilis jalapa Allamanda catiartica Plumiera rubra Bryophyllum pirnatum Polypodium scolopendria Calophyllum inophyllun Polyscias guilfoylei Canavalia microcarpa Pseuderanthemum laxiflorum Cassytha filiformis P. carruthersii Catharanthus roseus Psilotum nudum Cordia subcordata Russelia equisetiformis Cordyline terminalis Scaevola sericea Crinum asiaticun Sida fallax Cyperus javanicus tachytarpheta jamaicensis Dodonaea viscosa Thespesia populnea Eragrostis amabilis Triumfetta procumbens Euphorbia heterophylla Wedelia biflora Guettarda speciosa 5. Leaves are used in composts prepared for babai cultivation Boerhavia diffusa Sida fallax Cordia subcordata Thespesia populnea Ficus tinctoria Triumfetta procumbens Guettarda speciosa Wedelia biflora Hibiscus tiliaceus 6. Fiber plants used in skirts, fishing Lines, nets, etc. Agave sp. Cyperus laevigatus Cassytha filiformis Hibiscus tiliaceus Cyperus javanicus Tacca leontopetaloides ite Dyes and tanning materials Ficus tinctoria Morinda citrifolia Rhizophora mucronata 8. Plants used to scent coconut oil Dodonaea viscosa Polypodium scolopendria Fimbristylis cymosa Premna obtusiofolia Ocimum basilicum Rhizophora mucronata Pemphis acidula Terminalia catappa 9. Plants used to feed pigs and chickens Boerhavia diffusa : - Phyllanthus amarus Eleusine indica Portulaca lutea Ficus tinctoria Portulaca samoensis Fimbristylis cymosa =768- CHAPTER 9 VEGETATION AND FLORA PART I: VEGETATION With the exception of Pemphis stands and, of course,of the mangrove proper, no primitive vegetation types can be recognized today. The original formations have been so thoroughly modified by man that there is no trace left of them, especially as all these islands are rather densely populated. Besides the coconut palm, which covers the largest area, some species may still have a certain density and show marked preferences for some habi- tats, but they form a secondary vegetation rather than ruins of former original types. In many cases the primary components are now represented only by iso- lated specimens, which tend to disappear not only because of the growing prevalence of the coconut palm but also because the natives do not care to preserve or propagate them. This is the case for trees such as Pisonia, Cordia, etc. In some islands even valuable trees such as pandanus may be disappearing, as they are considered of minor importance in comparison with the commercial value of copra. The Gilbert Islands offer today the following vegetation types: | I. Vegetation of the seaward side Scaevola, more less dense, forms an almost unbroken belt. Their density is always greater than that of other neighbouring plants. This species grows nearest to the shore. Messerschmidia in isolated specimens or small groups are found, most of them with twisted trunks. Their height does not exceed 3m. Pandanus, are more or less numerous, but always as isolated individuals. The prevailing low plants are: Lepturus, which is very seldom found in the shaded central zone, and Fimbristylis. A few Triumfetta and Cyperus sometimes grow among them. The coconut palm begins in this zone, the first rows growing a little above high-water mark. II. Vegetation of the interior The area occupied by other species is usually conditioned by the density of the coconut palms. If these are very close together, we find only the following low plants: Thuarea, Fimbristylis, Euphorbia and gramineae of the genera Stenotaphrum and Cenchrus, with a few small Scaevola here and there. If the coconuts are less dense, there will be some tree species such as: Pandanus, Guettarda and Morinda; and in addition to the low plants already mentioned, Boerhavia, Triumfetta, Fleurya, Sida, Dodonaea and Cassytha. The grasses are almost all present and Psilotum and Polypodium occur here and there at the foot of the coconut trees. Of course the number of these plants and the area they cover vary with the degree to which the ground is tended. O79. If the coconut palms are very sparsely planted, some trees such as Cordia and Pisonia may occur, though rarely nowadays. Many low plants survive with difficulty in open areas while species such as Sida fallax, which seem to prefer strong sunlight, can achieve great extension there. ITI. Vegetation of the lagoon side(area of roads and villages) The roadsides being generally cleared to a certain width, few plants are to be found there except common grasses. On the other hand, the shrubs and trees of the interior may be found again between the cleared road zone and the edge of the lagoon, with the addition of small groups of Messerschnidia forming a narrow strip slightly in front of the coconut palms (Pilate X1Va). One plant will be found growing densely on sandy areas that are always damp below the surface through tidal seepage; this is Fimbristylis, which tolerates high salinity. On the contrary, Scaevola, Guettarda and Lepturus always grow above the level of the highest tides. Village areas offer a very different aspect, due to the number of plants cultivated in their immediate vicinity and around the houses. Among food trees, in addition to coconut palms which are widely spaced, breadfruit trees predominate, and sometimes reach a large size. Pandanus trees are found in varying numbers, mostly around the village periphery, except in the southern islands where they are given the same choice locations as Artocarpus. Papaya trees are found in every village and are often very tall. Banana plants are sometimes a component of this vegetation, but are grown only in pits. Small pumpkin patches (Plate XIVc) are seen around the houses in southern islands, generally side by side with numerous tobacco plants (figure XVc), while Ficus tinctoria is usually found a little behind the last houses. Tomatoes and sweet potatoes are very scarce and we saw them only on a few islands in the centre and north of the Gilbert Group. One of the characteristics of the Gilbertese village is the great variety of ornamental plants. The most commonly found are Crinum, Russelia, Mirabilis, Catharanthus and Pseuderanthemum. The low plants considered as weeds vary in abundance, of course, according to the cleanliness of the village and are, in fact, rather scarce. Euphorbia prostrata, Fimbristylis, Phyllanthus, Eragrostis and Digitaria are most often seen. In addition to village and roadside vegetation, the Pemphis type should be given special mention. This shrubby plant (Pemphis acidula Forst.) forms thick stands, often spreading over large areas, just at the limit of the highest tides of the lagoon, and above the first depressions filled by high tides and occupied by Rhizophora (Plate XIVd). The latter may often cover large areas which are submerged at high tide. -80< PART II; PLANTS COLLECTED IN THE GILBERTS None of the plants growing in the Gilbert Islands are endemic. In fact, most of them belong to the general coral island flora in the tropical Pacific. To assess the time when they first appeared, even very approxi- mately, is a most difficult task. It seems, however, that three main time divisions may be recognized. I. Plants present on the islands before the arrival of the first occupants It is logical to assume that this group can include only plants with seeds likely to resist a prolonged stay in salt water and retain for a long time their germinating powers. These may have been introduced to these islands by currents and trade winds, and they would have been the first plant stock of these territories. Other carriers such as birds, may have (1) brought very tiny seeds (gramineae for instance), but to a very small extent. II. Plants introduced by successive waves of immigration amet logical to place in the second group the ancient food plants. The natives carried them along during their migrations. They probably also brought plants of ritual significance. III. Plants introduced since the arrival of Europeans The third group includes only plants introduced less than 200 years ago by the first navigators, by the traders who followed them, and finally in quite recent times by Europeans residing permanently or temporarily in these islands, missionaries, government officers, etc. To these intentionally- introduced plants should be added those which arrived by accident. The natives also took part in these introductions and still do so at every op- portunity. Taking into account the above considerations, we shall therefore say: Group I for plants of preehuman introduction Group II for plants of ancient introduction Group III for plants of recent introduction It will be easily understood that, detailed historical documents not being available, the limits between these groups have to be somewhat flex- ible (especially between groups I and II). (q) The extent of such bird transport must be larger than stated here, as plants with sticky and spiny fruits, such as Pisonia, Boerhavia and Tri- unfetta, and fleshy fruits, such as Scaevola, are known to be carried around by birds. = Ed. Ole A list of plants collected, with brief notes on each, appears below. It should be noted that the vernacular names quoted are those given us at the collecting locality, and that they may vary from one island to the next. A set of the specimens is deposited in the herbarium of the Museum d'Histoire Naturelle, Paris.(1 It was impossible to collect again a botanical specimen of each species in each territory visited, since we had to attend to many other more important matters. This task would not, however, have been useless since the natives always showed an understandable hesitation when confronted with dried plants in the herbarium. Moreover, in some islands we were unable to stay long, and no collections could be made, so that our herbarium does not claim to be ex- haustive. We feel that a botanist could still find a number of plants especially among very recent introductions. These are still continuing. When the Gilbertese return from working abroad they generally bring back new plants. A remarkable fact is that they almost always choose plants with either ornamental leaves or fragrant flowers which may be fashioned into the leis and garlands required for the dances and songs that form so important a part of the daily life in these islands. Singing and dancing are a real need for these people, who would have little joy in life without them. We have not, of course, referred to uses of no economic significance such as those connected with games, magic,etc. Finally, most of the crypto- gams do not appear after the phanerogams listed, since the identifications awaited from Professor R. Heim of the Paris Museum have not yet reached us. The algae from fish-ponds were identified by Dr. Francis Drouet, see Ap- pendix I, p. 187. Dr. Katherine Luomala, of the University of Hawaii, very kindly pro- vided us with a copy of her manuscript on ethnobotany of the Gilbert Islands, including a history of plant collecting in the Gilbert Islands. Since that time it has been published as Bulletin 213 of the Bernice P. Bishop Museum, 1953. Material included in it has deliberately not been repeated here, as the original may be consulted. List of plant species Abutilon indicum (L.) Sweet Malvaceae Group III. Gilbertése name: te kaura ni banaba; English name: Indian mallow. Locality: Tarawa, no. 124; Ocean I., 124 bis. This is very rare in the Gilberts, seen only on Bikenibeu, where a few specimens had been planted as ornamentals in front of a native house. It is (I) Through the courtesy of the authorities of the Museum d'Histoire Natur- elle, Paris, it has been possible to reexamine specimens of most of the plants numbered 1-139. Changes in the nomenclature of certain of these have been incorporated into the following list to bring it into accord with the names generally in use in recent papers on Pacific coral atolls. Where there seems some room for disagreement about the group to which the plant has been assigned, the suggestion of a different placing has been indicated in parentheses. - Ed. a Shek very common on Ocean Island (Banaba) whence it was doubtless brought, as in- dicated by the Gilbertese name. Acacia farnesiana Willd. Leguminosae. Group Lit: English; Spring aroma Locality: Nikunau, no. 154 Medium-sized shrub, very rare in the Gilberts, seen only in a village on Nikunau. Uses: The very fragrant flowers are used in garlands. Acalypha amentacea var. Euphorbiaceae, Group IIil Gilbertese: te aronga. English: copper leaf. Locality: Abaiang, no. 78 This ornamental shrub, with red leaves, may be over 2 m tall; it is found on all the islands, mostly in villages. Uses: The leaves are used in garlands. Acalypha amentacea var. Euphorbiaceae. Group Jif. Gilbertese; te aronga Locality: Abaiang, no. hi, A shrub 1.5 to 2m tall, with brown-purplish leaves; planted around houses and villages. Uses: The leaves are used in garlands. The Gilbertese do not eat the young shoots or leaves which are cooked as vegetables in other areas of the Pacific. Acalypha amentacea var. Euphorbiaceae. Group III. Locality: Tarawa, no. 115; Nauru, 115 bis This very rare plant was seen only at the Residency at Bairiki, and on Beru, where it was about 1m tall. It is very common on Nauru, where it forms hedges, and whence it was probably brought by Gilbertese employed in the phosphate works. Adiantum cuneatum L. & F. Polypodiaceae. Group III. Locality: Bairiki, Tarawa, no. 111 This fern about 4O cm tall is cultivated in pots on verandas of Euro- pean houses, Agave sp. Amaryllidaceae. Group III. Gilbertese: te robu (English: rope) English: green aloe Locality: Betio, Tarawa, no. 119 Only a few plants seen in Tarawa, Abaiang, Arorae and Nikunau. Uses: After retting, excellent fishing lines can be made from the fibers. WB3u Allamanda cathartica L. Apocynaceae. Group III. Locality: Tarawa, no. 117 | A liana with large yellow flowers, very rare, seen only in garden at Bairiki, Tarawa and in Beru. Very ornamental. Amaranthus dubius Mart. Amarantaceae, Group IIT. Locality: Tarawa, no. 9h This weed, 30-40 cm tall, with light green inflorescence, was seen only on Betio Islet. It is one of the food plants of Ethmia colonella Wals. Angelonia angustifolia Benth. Scrophulariaceae. Group III. Locality: Bairiki, Tarawa, no. 137 A blue-flowered plant, not over 35 cm tall, found only in European gardens. Antigonon leptopus H. & A. Polygonaceae. Group III. English; coral vine Locality: Tarawa, no. 108 Climber, with old rose flowers, seen only in the garden of the Resi- dency at Bairiki. Artocarpus altilis (Park.) Fosb. Moraceae. Group II. Gilbertese: te mai (general name) Locality: Mission, Butaritari, no. 85, 86; Tarawa, no. 122; Betio, Tarawa, NOs. L336 Nos. 85 and 122 represent variety te bukiraro; no. 86, te mai kora; no. 133, te moti ni wae. Artocarpus mariannensis Trec. Group II. Locality: Tarawa, no. 121 This is variety te maitarika. Artocarpus altilis X mariannensis Group II. Locality: Tarawa, nos. 120, 123 No. 120 represents variety te mai; no. 123, variety te ang ni Makin. For a detailed study of Artocarpus, see chapter 6. Asclepias curassavica L. Asclepiadaceae. Group III. Locality: Tarawa, no. 151 Very rare in the Gilberts, seen only once at Bikenibeu and once in the village at Betio. The larva of Danais plexippus usually feeds on this plant, and the in- sect is very rare in the Gilberts, only one was seen in the space of six months, in the Residency garden in Bairiki. Asplenium nidus L. Polypodiaceae. Locality: Tarawa, no. 112 Seen only as cultivated fern in a European house, Bairiki. -Ohe Bacopa monnieri (L.) v. Wettst. Scrophulariaceae. Group III. Locality: Butaritari, no. 70 A small plant, not over 40 em tall, flowers white. Seen only on the edges of the fish ponds. Uses: The natives do not seem to use it, but the missionaries are said to eat it in salads, like cress. Bambusa sp. Locality: Marakei, no. 75 Robust and erect grass, stems used as fishing rods for very small fish. Barringtonia asiatica (L.) Kurz Barringtoniaceae. Group I. Gilbertese: Te baireati Locality: Tarawa, no. 10, Very rare in the Gilberts, only a few trees grow on Tarawa, Little Makin and Butaritari. Evidently there were no more at the time when Grimble wrote: "one or two Baireati trees grow on Butaritari and Little Makin but the supply of seeds is obtained mainly from the western beach of any island, where it is sometimes washed ashore in considerable number during the season of westerly gales." Uses: It seemed to us that few natives knew its use as fish poison to catch fish on reefs. Boerhavia diffusa L. Nyctaginaceae. Group I. Gilbertese: te wao Locality: Tarawa, no. 13; Abemama, no. 13b. Herbaceous plant with tuberous root, very common everywhere. When ex- posed to intense light the leaves turn a purple color. | Uses: In cases of prolonged drought, the plant may be used for food. Merrill (1943, p- 55) reports that the roots are eaten in Fiji in times of scarcity, but affect the kidneys and should be used with caution. Generally in the Gilberts, this plant is fed to pigs. In the treatment of a disease called te ba (rickets ?), leaf petioles are crushed and boiled in toddy. The leaves are used in compost for babai plants. Boerhavia is one of the food plants of the larva of Prodenia litura (F.). Boerhavia tetrandra Forst. Nyctaginaceae.. Group I. Gilbertese: te wao Locality: Bukenibeu, Tarawa, no. 27 le By Creeper, rather common, with small mauve flowers; can live on poor ground if there is enough shade. No uses known. Bougainvillea spectabilis Willd. Nyctaginaceae. Group III. Locality: Tarawa, no. 143 Liana, uncommon in the Gilberts, and not well developed. Seen only in gardens of the missions and of the Residency at Bairiki. Bruguiera conjugata (L.) Merr. Rhizophoraceae. Group I. Gilbertese: te tongo; te buangi Locality: Abemama, no. 92 Sometimes occupies important areas, especially on Abemama where it forms very dense stands on the sandy zone along the lagoon. Varies in height between 4 and 5 m, with a diameter seldom over 15 cm. Uses: The wood is used in building houses (roof props). -Bryophyllum pinnatum (Lam.) Kurz Crassulaceae. Group III. Gilbertese; te ang English: life-plant Locality: Bikenibeu, Tarawa, no. 21 Erect plant 50-60 cm tall, leaves fleshy; rather common. Uses: The flowers are used in garlands. In cases of headache, leaves are crushed in water and liquid used to wash the head and make Peek which can be renewed as necessary. ; Calophyllum inophyllun L. Guttiferae Group I (or II?). Gilbertese: te itai. English: punai nut (tamanu in Polynesia) Locality: Teaoraereke, Tarawa, no. 41 This is one of the most beautiful trees in the Gilberts, 6-7 m tall usually, ‘but occasionally up to 12 or 15 m with a diameter of 60 cm at the base, with a straight trunk and no low branches. This tree must have been . much more abundant formerly and its present scarcity is much to be re- gretted. Uses; The flowers are used in garlands; the fruits in children's games. The wood is excellent, hard, resistant and fine-grained, and is used in canoes, especially for the bow piece and ribs. It is also used for diving goggles. In Tahiti (Papy 1951), oil extracted from the fruits is used in pharmacy and exported, but this utilization is unknown in the Gilberts. -86- Canavalia microcarpa (DC) Piper Leguminosae Group II or III (or I) Gilbertese; te kitoko (sometimes te ruku, general name for liana) Locality: Tarawa, no. 96 Growing luxuriantly on Betio Islet and in the garden of the Residency at Bairiki; poorly developed on Arorae. Uses: The roseemauve flowers are used in garlands. The natives might be taught the possible use of Canavalia as a cover plant. Canna indica L. Scitamineae. Group IIIf. Locality: Tarawa, no. 144, Seen only in gardens of European houses, and usually yellow with chlorosis or lack of water. Capsicum frutescens L. Solanaceae. Group III. Gilbertese: te beneka (from vinegar) English: chili pepper Locality: Tarawa, no. 39 Shrubby plant 80-100 cm tall, very rare, seen only on Eetio. Uses; Used as condiment by Europeans, but not by natives. Carica papaya L. Caricaceae. Group III. Gilbertese: te papaya, or te mwemweara (according to Luomala). English: pawpaw, papaya. Locality: Abemama. Papaya trees are rather abundant, but only in villages. With a few ex- ceptions, such as those in missionary gardens, they receive no care. They may reach a height of 5 m or more, but produce few fruits. The Gilbertese, except for children, do not seem to care much for papayas, although the better cultivated ones have an especially good flavor. When eaten green, the fruit are cooked in coconut milk. Cassia occidentalis L. Leguminosae. Group IIT English: nigger coffee Locality: Betio Islet, Tarawa, no. 53 Very rare, seen only on Betio, but seeming to increase rapidly. The natives believe its introduction to date from American landings. No known uses. Cassytha filiformis L. Lauraceae. Group I. Gilbertese: te ntanini Locality: Bikenibeu, Tarawa, no. 11 -87- Climbing parasitic plant which may sia cover shrubs and trees, in clearings and open places. Uses: The slightly acid fruits are eaten, especially by children. The dried and decorticated stems are sometimes used for riri.., Fresh stems are used for stringing flowers into garlands. Cassytha is one of the plants on which the larva of Prodenia litura (F.) lives. Casuarina equisetifolia L. Casuarinaceae Group III English: Australian pine, iron wood. Locality: Butaritari, no. 65. Only 3 Casuarina trees were seen in the whole area visited. Two were young plants 1.8 m tall in the garden of the Residency at Bairiki on Tarawa, and had been planted in July 1951 (brought from Australia). The third, said to have been planted by Americans during the war is on Butaritari and is a straight fine tree about 10 m tall ard 30 cm in diameter, about 7 years old. -It.grows near a group of houses. It is surprising that this tree is not more abundant on these islands,as on so many others in the Pa- cific. It could profitably be planted on shores not used for coconuts. The chart for Aranuka notes a stand of Casuarina. Actually this is a forest of exceptionally tall and dense Rhizophora. Catharanthus roseus (L.) G. Don Apocynaceae Group III Gilbertese: te buraroti. English: Madagascar periwinkle Locality: Bikenibeu, Tarawa, no. 38 Small shrub, 30-50 cm tall, with pink or white flowers, seen only in villages around houses. When occasionally found in the jungle, it indi- cates the former site of a tomb or village. Uses: flowers are much prized for especially fine garlands. Cenchrus echinatus L. Gramineae Group II Gilbertese: te kateketeke Locality: Bikenibeu, Tarawa, no. 32 Not very common, in fact aimost impossible to find before the rains. The natives seem to destroy it whenever possible, apparently following in- structions given a long time ago to check the rapid spread of this grass. Uses: young plants are eaten by pigs. -88- Chloris inflata Link Gramineae Group III Locality: Tarawa; Abemama, no. 88 Not very common; grass 15-40 cm tall, with mauve inflorescences. Probably introduced by American troops. Clitoria ternatea L. Leguminosae Group III Locality: Tarawa, no. 113 Seen only in the garden of the Residency at Bairiki. Royal blue flowers. Cocos nucifera L. Palmae Group I or II (prob. IT) Gilbertese: te ni. English: coconut paln. Fragments of leaves and inflorescences collected on Tarawa, but unnumbered. See Chapter 4 for a full discussion of this tree. Colocasia esculenta (L.) Schott Araceae Group III (or IT) Gilbertese: te taororo. English: taro. Locality: Bairiki, Tarawa, no. 103 Taro is little cultivated in the Gilberts. A few plants are found here and there in the babai pits, but are always poorly developed, as they do not stand well the conditions under which babai thrive. Taro tubers become rotten before 8 months have passed. On Betio, where we saw more than else- where we learned from the owmers: that taororo do not grow well in babai pits unless dirt is piled at the bottom of the pit to raise the plant suf- ficiently above water; that without this precaution, the tubers would start rotting after 6 months; that, however, if the ground is not too wet, tubers weighing about 750 g may be obtained, although tubers pulled up when 6 months old are too often already rotten to a depth of a few cm; that it is not necessary to build composts as for babai. Taro is much less appreciated than babai by the Gilbertese. Some like it, but say it is too bland and not "heavy" enough, meaning probably that it is tasteless and does not seem to fill the stomach. Another objection comes from the fact that taro, since it must be pulled up after a few months, does not form like babai a reserve which can be tapped when neces- sary. Besides, several natives told us that the reason they did not care for taro was that it cannot be kept for a long time as babai is in the prepared kabuibui ni babai. Most of the taros seen were more or less badly attacked by the larva of Prodenia litura (F.). -89- Cordia subcordata Lam. Boraginaceae Group I Gilbertese: te kanawa Locality: Bikenibeu, Tarawa, no. 28b, 84; Abemama, no. 28 Not very common, but must have been more abundant in the past. Shrub forming small bushy groups, or growing as isolated individuals, well formed, up to 5m tall. Appears to be more common along shores, but may have been more abundant in the interior of islets before coconuts were planted there. Its increasing scarcity is much to be regretted, as the wood is valuable. Uses: Flowers are much appreciated for garlands. Leaves are added to babai composts. Twigs are used as medicine: the bark is peeled off and the stems scraped. The scrapings are squeezed with water (rain water if possible) by hand. The liquid is given to drink to people afflicted with persistent diarrhoea. The wood is mostly used for canoes, especially for the key piece of bow and stern (te karetaba). Also used in fishnet floats. When the plant is shrubby, it provides good fishing poles. Cordia is one of the food plants of the larva of Ethmia colonella. Cordyline terminalis (L.) Kunth Liliaceae © Group III Gilbertese: te rauti pete Locality: Butaritari, no. 43 Shrubby plant about 2m tall, with pale purple flowers, grown as orna- mental in villages around houses; seen only in Tarawa and Butaritari. Crinum asiaticum L. Amaryllidaceae Group II Gilbertese: te ruru, te kiebu. Locality: Teaoraereke, Tarawa, no. 42 Cultivated in villages, where flowers are much appreciated. Also often marks the site of ancient graves. sean height of 50 cm. Uses: Flowers are used in garlands. The young inner leaves are crushed and the juice squeezed out and poured in the ear in cases of earache. Cucurbita pepo L. Cucurbitaceae Group III Gilbertese: te baukin. English: pumpkin Locality: Bairiki, Tarawa, no. 136 Cucurbits are grown in most islands. There appear to be both squashes and pumpkins. In some villages in the south islands, the plants cover most of the space between houses (Plate XIVb). =O Cycas circinalis L. Cycadaceae Group III English: sago-palm, Cycas 5 Locality: Tarawa, no. 147 Very rare in the Gilberts, some fine specimens not over 2.50 m tall were seen at Butaritari and in the garden of the Residency at Bairiki, Tarawa. Cyperus compressus L. Cyperaceae Group III Locality: Tarawa, no. 99 Plant 20 cm tall, inflorescences pale green. Seen only on Betio Islet. Cyperus javanicus Houtt. Cyperaceae Group II (or I) Gilbertese: te ritanin Locality: Bikenibeu, Tarawa, no. 20 Plant 50-60 cm tall, found mostly as a weed in poorly tended babai pits. Uses: Inflorescences are used in making garlands, stems in riri. Cyperus laevigatus L. Cyperaceae Group Lf or FE Gilbertese: te maunei Locality: Nikunau, no. 129 Sedge not over 50 cm tall, with brown inflorescences, growing on the edge of the lakes, with water coming up to 1/3 of its height. Uses: Used in riri. The stems can be pulled off without uprooting the whole plant. Cyperus polystachyos Rotto. Cyperaceae Group III Locality: Betio, Tarawa, no. 98 Plant 30-45 cm tall, inflorescences pale green tending toward yellow. Very common on Betio. Cyperus rotundus L. -Cyperaceze ~ Group II or III (probably III) Locality: Tarawa, no. 100; Nikunau, no. 131b Plant 30 em tall, inflorescences dark red. Very common on Betio, seen also on Nikunau, where it is rare. aO]- Cyrtosperma chamissonis (Schott) Merr. — Araceae Group apt Gilbertese: te babai Locality: Nikunau, no. 127 See chapter 7. Datura metel L. Solanaceae Group III Gilbertese: te urintiana English: thorn-apple Locality: Tarawa, no. 81 Shrub with mauve flowers, up to 1.20 m tall, found in villages as ornamental. Seen only on Tarawa, Butaritari and Abaiang. Delonix regia (Boj.) Raf. Leguminosae Group III Gilbertese: te tua English: flame tree Locality: Tarawa, no. 149 Flame trees are very rare in the Gilberts, seen only in some villages of Tarawa, Beru and Aranuka, as isolated specimens. Only those on Aranuka seemed normally developed. Digitaria pruriens var. microbachne (Presl) Fosb. Gramineae Group III Gilbertese: te uteute (means grass) (or II) Locality: Tarawa, no. l3la, 132 Grass not over 50 cm tall, with grey-green inflorescences, seen only on Tarawa and Abemama, uncommon. Dodonaea viscosa (L.) Jacq. Sapindaceae Group II Gilbertese: te kaiboia Locality: Tarawa, no. 60 Shrub rather common in the Gilberts, 3 to 6 m tall. Grows mostly on the site of former villages, probably because numerous seeds were scattered there while garlands were being made. Uses: Fruits are used in garlands. Young leaves are used to perfume coconut oil. Stems give good fishing poles, and also frames for dip nets. Eleocharis geniculata (L.) R. &S. Cyperaceae Group III (or II) Gilbertese: te maunei Locality: Bikenibeu, Tarawa, no. 37 Plant 8-10 cm tall, never seen outside babai pits. The same name, te maunei is applied to Cyperus laevigatus L. EQp- Eleusine indica (L.) Gaertn. Gramineae Group III Gilbertese: te uteute (grass) Locality: Bikenibeu, Tarawa, no. 35; Abemama, no. 89 This rather common grass is eaten by pigs when young. Eragrostis amabilis (L.) W. & A. Gramineae Group III Gilbertese: te uteute te aine (female grass) Locality: Bikenibeu, Tarawa, no. 36 Very common grass, found more in the interior, along roads, than along shores. Uses: Inflorescences are used in garlands. Children also attach them to sticks to chase flies. Euphorbia chamissonis Boiss. Euphorbiaceae Group I Gilbertese: te tarai Locality: Bikenibeu, Tarawa, nos. 8, 2k. Very common on all the islands; plants 30-40 cm tall. Uses: The plant is used as a purgative. The latex is also collected in a shell of te koikoi (Tellen scobinata L.), mixed with a little maimoto (coconut water), and the mixture given to people saved from drowning, to make them throw up the water absorbed. This species is one of the food plants of the larva of Prodenia ditura (RA). Euphorbia heterophylla L. Euphorbiaceae Group III Gilbertese;: te kabekau Locality: Betio, Tarawa, no. 55 Uncommon plant, 35-40 cm tall, very recently introduced by American troops. Uses: Leaves are used in garlands. Euphorbia hirta L. Euphorbiaceae Group III Gilbertese: te tarai Locality: Teaoraereke, Tarawa, no. 83 Plant about 4O em tall. Euphorbia prostrata Ait. Euphorbiaceae Group III Locality: Bikenibeu, Tarawa, no. 23 Low creeping plant, 10-30 cm across, branched, leaves dark red. Very common around villages in almost all islands including Nikunau, Beru, and Arorae where American troops did not go. The name "te amerika" used in Tarawa, and implying introduction by American troops therefore does vot seem justified. Ficus carica L. Moraceae GroupiLinvs: Gilbertese: te biku. English; fig Locality: Abemama, no. 150 Fig trees are very rare in the Gilberts, and are found only in Mission gardens. They develop well and their fruit are excellent. The Gilbertese could probably obtain good fruit by giving the trees the same care as to breadfruit. Ficus tinctoria var. neo-ebudarum (Summ.) Fosb. Moraceae Group I (or II) Gilbertese: te bero Locality: Bikenibeu, Tarawa, no. 3 Trees 4-7 m tall, about 40 cm around at 1 m above the ground; in the southern islands, they are smaller and the trunks not over 25 cm around. It is a cultivated plant. Each landowner has'a few plants on his land. Reproduction is usually by cuttings, spontaneous seeding is rare. Uses: The fruits are picked when ripe, sometimes when green, cooked, mashed with a pestle or in a meatgrinder. The puree can be eaten after being sweetened with kamaimai or sugar, or preserved for future use. Ii is then called kabuibui ni bero and is prepared as follows: the puree is spread thinly on wide Guettarda leaves spread on coconut leaf mats, and left to dry in the sun for 3-4 days. The Guettarda leaves are changed everyday as long as the paste remains sticky, later it is placed on a suit- able surface, coconut leaf mat, board, to complete drying. When dry, ka- buibui ni bero is stored in tin cans and will keep for several months. The fruits are also used, though much less now than formerly, to dye various objects such as hats, mats, etc. This dye was preferred to te non (Morinda citrifolia). | Sis Rg The leaves are fed to pigs, sometimes used in composts for babai or breadfruit. Branches are sometimes used in building houses, especially as supports for roof thatching. Fimbristylis cymosa var. Cyperaceae Group II or III (prob- Gilbertese: te uteute te mane (male grass) ably I) Locality: Bikenibeu, Tarawa, no. 9 The most common herb on the islands, and the most resistant to drought. Forming tufts. -9h. Uses: The juice squeezed by twisting the stems is used as eye drops in cases when a foreign body under the eyelids causes pain. The roots are used to perfume hair oil, and formerly, riri. The chickens are very fond of the seeds. Fleurya ruderalis (Forst.) Gaud. ex Wedd. Urticaceae Group II (or I) Locality: Teaoraereke, Tarawa, no. 56 Low plant about 20 em tall, not common. Always considered as a weed of coconut groves. One of the food plants of the larva of Hypolimnas bolina var. rarik (Esch.) Gaillardia pulchella Foug. Compositae Group III Locality: Bairiki, Tarawa, no. 118 Small plant about 30 em tall, ray flowers yellow, disc flowers golden brown. Seen only in gardens of Europeans. Gliricidia sepium (Jacq.) Steud. Leguminosae Group III Locality: Betio, Tarawa, no. 95 Pink-flowered tree, seen only on Betio and Bairiki (Tarawa). On Betio, only a few specimens less than 3 m tall, which mark the site of the worthy but unsuccessful plant introduction garden initiated by Administrator Wil- liams (1946). On Bairiki, one tree at least 4 m tall and well developed grows in the garden of the Residency. It is placed in exceptionally good conditions and this species may not be able to become more widespread. Gomphrena globosa L. Amaranthaceae Group III Gilbertese: te motiti. English: Bachelor button Locality: Betio, Tarawa, no. 138 Plant 30-40 cm tall, with flowers and bracts of a purplish red. Rare, seen only on Beru and Tarawa in front of native houses. Gramineae, undetermined | Locality: Abemama, no. 90 Grass 85 cm tall, sterile specimen, undeterminable. Guettarda speciosa L. Rubiaceae Group I Gilbertese: te uri Locality: Bikenibeu, Tarawa, no. 25 Very common tree up to 6 m tall, one of the main components of atoll vegetation, especially in the center of islets. Its dead leaves are im- portant in the formation of surface soil. -95- Uses: The wood is used in houses for rafters and wall frames, in canoes for hull and ribs. Formerly used in fire-making by friction. Flowers are commonly used in making garlands. Leaves are used as ferti- lizer for babai and other plants, alone or mixed with other material. All pastes and preserves are spread on them for sun-drying. They have other domestic uses, as dishes and on various occasions when we would use paper. The soil (te iarauri, te ianuri) scraped under Guetiarda trees is added to composts for babai and other cultivated plants. Larvae of Chloauges woodfordii Butler and Cephonodes pictus Cramer, feed on Guettarda. Hedyotis biflora (L.) Lan. Rubiaceae Group III Gilbertese: te uteute (herb) Locality: Butaritari, no. 67 Hernandia sonora L. Hernandiaceae Group I Gilbertese: te nimareburebu Locality: Teaoraereke, Tarawa, no.48 Tree about 4 m tall, rather rare, seen only on the lagoon side. Uses: The wood is used in building houses, and for outrigger floats. Hibiscus rosa-sinensis L. Malvaceae Group III Locality: Tarawa, no. 145 Uncommon shrub, seen as isolated specimens in front of native or European houses. Foliage poorly developed, often chlorotic. The flowers often fall before opening, especially the double ones. Hibiscus tiliaceus L. Malvaceae Group I (or II) Gilbertese: te rao (Abaiang); te kiaiai (Beru) English: sea coast mallow Locality: Abaiang, no. 79 Tree 4 to 6 m tall, uncommon, completely absent from some islands; only in the south, and especially in Beru, is it seen in some numbers. It is hard to know whether it has never been more abundant, or has decreased in importance, like Cordia, Calophyllum and other trees. Its scarcity is much to be deplored as it has many diverse uses. Uses: The retted fibers are sometimes used to make riri. The branches are sometimes used in outrigger booms, because of their elasticity. The sprouts, when straight, make good fishing rods. The flowers are worn by dancers. The leaves are much appreciated for fertilizing babai pits. Large ones way replace Guettarda leaves as food wrappings. -96- Ipomoea batatas (L.) Poir. Convolvulaceae Group II (or III) Gilbertese: te kumara English: sweet potato Locality: Tarawa, no. 155 Sweet potatoes are rare in the Gilberts, only a few timid attempts to grow them can be observed here and there. Yet satisfactory results could be obtained in years of normal rainfall. Ipomoea pes-caprae (L.) Roth Convolvulaceae Group II (or III) Gilbertese; te ruku English: Beach morning glory Locality: Tarawa, no. 152 Uncommon, seen only in some abundance in Betio and near the landing place at Bairiki. Ipomoea tuba (Schlecht.) G. Don Convolvulaceae Group I Gilbertese; te ruku English: moon flower Locality: Tearinibai, Tarawa, no. 57 Very common liana, very abundant especially in damp places near lakes and ponds, climbing high into the trees. Uses: Stems and leaves are crushed in water, and used as shampoo. To kill lice, the shampoo is used in concentrated form. This plant is one of the food plants for larvae of Chromis erotus (Cramer) and Protoparce convolvuli subsp. distans Butler (Sphingidae). Ixora coccinea L. - Rubiaceae Group III Locality: Bairiki, Tarawa, no. 106 Very rare plant, with green leaves turning dark red, and scarlet flowers. Seen only in the garden of the Residency at Bairiki and in a mission garden on Abaiang. Jussiaea suffruticosa L. Onagraceae Group II Gilbertese;: te mam English: willow primrose Locality: Teaoraereke, Tarawa, no. 46 Found only as a weed, with Cyperus javanicus, in the poorly tended . babai pits. Not very common; about 40 cm tall. Lantana camara L. Verbenaceae Group III Locality: Bairiki, Tarawa, no. 107 Uncommon, only a few plants seen on Betio (orange flowers) and at Bairiki (pink flowers), 30 cm to 1 m tall. Uses: Flowers worn in hair. -97- Lepturus repens R. Br. Gramineae Group I Locality: Bikenibeu, Tarawa, nos. 30, 33, 34 One of the most common grasses, extremely abundant; growing mostly on the edges of islets rather than in the center. Forming tufts near the shores and creeping plants further in the interior. No. 30 represents variety subulatus Fosb.; no. 34 variety maldenensis F. Brown; no. 33 is a sterile specimen not determinable to variety. Lumnitzera littorea (Jack) Voigt Combretaceae Group I Gilbertese: te aitoa Locality: Butaritari, no. 80 Shrub 4-5 m tall, seen only in Butaritari where many plants grew, often in dense stands near the swamps and lakes converted into fish ponds. Uses: The scarlet flowers are much appreciated for garlands. The very solid wood is used in houses and for fish traps (te inai), because it does not deteriorate in sea water. Mangifera indica L. Anacardiaceae Group III Gilbertese: te mgo, te manko Locality: Onotoa, no. 142 Mango trees are extremely rare, and it is surprising that any can sur- vive and even produce some fruits, as did one in Butaritari. In Onotoa, a tree about 4 m tall had been planted 3 years earlier, as a sapling brought from Ocean Island where seedlings are common under old trees. It had not yet flowered. Another had died in the recent drought. Messerschmidia argentea (L.f.) Johnst. Boraginaceae Group I Gilbertese: te ren Locality: Bikenibeu, Tarawa, no. 15. Funafuti, no. la. Extremely common, as scattered groups of trees or forming a strip of sparse vegetation along the beaches, with similar density on lagoon and ocean sides. Average height 1.5 to 2.5 m, but some plants can be up to 5 m tall and over 2 i around 1 m above ground. Branches thick and brittle, leaves covered with silvery hairs. Uses: The young leaves, crushed with a pestle are placed in a te ing (fibrous stipular sheath at base of coconut leaf) and in water, and the juice squeezed out. It is a febrifuge. Dose for children, 3 leaves; for adults, 9 leaves. The wood, moderately hard, is the most appreciated as fuel. It some- times replaces te itai (Calophyllun) for the main piece of canoe bows. -93- Mirabilis jalapa Te Nyctaginaceae Group IIT Gilbertese: te aova. English: Four o'clock flower Locality: Bikenibeu, Tarawa, no. 1 Seen only in villages around houses. Said to have been brought from Ocean Island. Plant usually 40 cm tall; in Butaritari, plants with white, yellow or purplish flowers were seen. Uses: Flowers are used in garlands. Morinda citrifolia L. Rubiaceae Group II Gilbertese: te non Locality: Bikenibeu, Tarawa, no. 6 Present on all the islands and rather common. Plant 3-5 m tall. Flowers white. Uses: The wood is used to build houses, especially roofs. From the roots, a red dye is extracted and used to color mats, hats, etc. This ex- cellent stain, unfortunately, tends to be replaced by imported dyes, al- though the Handicraft Department tries to encourage anew the use of local dyes. The leaf-cutting bees Megachile cut most of their little discs from the edges of Morinda leaves, which are often so scalloped that the appear- ance of the tree is changed. The discs are rolled into tubes for the pro- tection of the larvae. Morinda is also the food plant of larvae of Chromis erotus (Cramer). a ine PRE Musa paradisiaca subsp. sapientum (L.) 0. Ktze. Musaceae Group III Gilbertese: . te’ banana The herbarium specimens (flower buds) of 3 bananas apparently of dif- ferent varieties, were lost during their transport from Tarawa to Ocean Island, and the photos of the plants were not enough for Professor L. H. MacDaniels to identify varieties. Bananas are not common in the Gilberts. They are usually grown in. pits, but not in flooded soil as for the babai. The banana plants are placed in slightly higher part of the pit, or in old pits partially filled .. up. At Butaritari we saw a few beautiful plants growing in a small de- pression behind the village near the last houses. The bunches were heavy with fruit. The most interesting observations on bananas were made at the Makin School on Abemama, and at the Manoku Catholic Mission on another islet of the same atoll. In a pit in the first locality, 30 banana plants showed a remarkable development; they had been planted 10 months previously, and one already had a flower bud. At Manoku the plants in some pits were already -99-= bearing, in others the plants were less vigorous and seemed weakened, pos- sibly as a result of serious damage caused by a Noctuid larva. Generally, it must be pointed out that the establishment of such banana plantations requires much work. A rather deep trench (about 1 m) must be dug around the bottom of the pit to keep coconut roots out. A layer of earth scraped from under Guettarda, Scaevola, etc. must be spread. Pieces of scrap iron must be buried here and there. Some school teachers have tried to fertilize the pits with algae and sea grasses but have failed, perhaps be- cause the plants were not washed in fresh water first. Father Brandstett of the Manoku mission said that if the suckers are removed while the bunch is developing on a banana plant, it ripens rapidly without growing larger. We asked some natives why they did not try to in- crease their banana production. Their answer was that the first bananas to mature would systematically be stolen and that, when the bunch was cut off, the owner would be deprived of his crop by bubuti. These customs are too deeply anchored in the Gilbertese life to be easily changed. Bananas are attacked by the larva of Prodenia litura. The leaves are often bored through before they unfurl. A single plant may bear hundreds of larvae. A beetle Pentodon sp. or Papuana sp. (see p. 28) burrows into the plant a little below soil level. Nephrolepis hirsutula (Forst.) Presl Polypodiaceae Group III Locality: Abaiang, no. 71 Seen also on Butaritari and Tarawa. Nerium oleander L. Apocynaceae Group III Gilbertese: te orian English: oleander Locality: Tarawa, no. 146 Uncommon shrub, not over 3 m tall, seen near churches on Tarawa, Arorae, Abaiang, etc. Nicotiana tabacum L. Solanaceae Group III Gilbertese: te kaibake English: tobacco Locality: Bikenibeu, Tarawa, no. 4 Tobacco is cultivated on practically every island, but is abundant only on the southern islands, especially Nikunau, where each house (Plate XIVc) is bordered with one or two lines of tobacco plants, 50 cm tolm tall. The care of plants and preparation of leaves are reserved for old women. Curing is rather sketchy and the resulting product far from satis- factory: the leaves are simply dried on hot stones. If more leaves are available than are needed daily, they are suspended singly or in bunches in the huts. The natives much prefer stick tobacco and use their own only when the store is out of sticks, as happens often, or when money is lacking. -100<- Ocimum basilicum L. Labiateae Group III Gilbertese: te marou English: basil Locality: Tarawa, nos. 105, 130 Rare plant, said to have been brought from Ocean Island, seen only on Tarawa and Nikunau as isolated individuals growing in front of houses. Flowers mauve, plant about 35 cm tall. Uses: The very fragrant leaves are much appreciated for garlands. Cultivated paln Palmae Group III Gilbertese: te bam (palm) Locality: Tarawa, no. 153 Uncommon, but well developed. Pandanus tectorius Park. Pandanaceae Group I Gilbertese: te kaina Locality: Teaoraereke, Tarawa, no. 50 see Chapter 5. Paspalum vaginatum Sw. Gramineae Group I Gilbertese: te uteute (grass) Locality: Butaritari, no. 72; Betio, Tarawa, no. 97; Nikunau, no. 123 Grass not over 30 cm tall, seen mostly in protected places on beaches and in swampy areas where it forms a very dense sod. Found in Butaritari along the fish ponds, in Betio near the beach, and in Nikunau in swampy places near the lakes, often with Cyperus laevigatus. Passiflora foetida var. gossypifolia ase ) Mart. Passifloraceae English: Stinking passion fruit : Group III Locality: Butaritari, no. 64 3 seen only /Butaritari, where it is not common. Uses: The fruits are eaten by the natives. Passiflora foetida var. hispida (DC.) Killip Group III Locality: Bairiki, Tarawa, no. 139 Pemphis acidula Forst. Lythraceae Group I Gilbertese; te ngea Locality: Aranuka, no. 16 -101- Extremely common on all the islands, in dense stands on sandy areas near the mangroves. Trees 2-4 m tall, sometimes up to 5 m with trunks over 0.5 m diameter at the base. Trunks irregular, irregularly branched, with cracked bark. Uses: The small fruits are sometimes eaten. The rotting wood added to coconut oil is used in a cosmetic. The young leaves are said to have anti- scorbutic properties, but are not eaten in the Gilberts. The roots, scraped in water, are used to stop hemorrage after child- birth; this drink is supposed to have real hemostatic properties. To cure running sores (te kiniman) a root is carefully cleaned, and the juice from the scraped bark applied to the sore. The dense and hard wood of Pemphis has many uses. Many pieces of the framework of houses are made of it. It is always used for the top part of canoe masts, and preferred to all others to make pestles. It is also used to make pipes, traps for moray-eels, and in former times, fishing hooks; it is not attacked by sea-water. Pemphis is one of the food plants of Achaea janata larvae. Pentas carnea Benth. Rubiaceae Group III Locality: Tarawa, no. 116 Seen only in the garden of the Residency at Bairiki. Plant 30-40 cm tall, with flowers of a very pale mauve. Phyllanthus amarus Sch. & Th. Euphorbiaceae Group II (or III) Gilbertese: te kaimatu Locality: Bikenibeu, Tarawa, no. 17 Very common low herb, much appreciated by chickens. Physalis angulata L. Solanaceae Group III Gilbertese: te bin (northern Islands), te baraki (southern islands); English: ground cherry Locality: Morikao, Abaiang, no. 66 Very common plant preferring sunny openings. Uses: The ripe fruits are eaten raw and much appreciated. Pilea microphylla L. Urticaceae Group III Locality: Bikenibeu, Tarawa, no. 87 Low plant, not over 6 cm tall, rather common on some islets such as Bikenibeu, especially near the lagoon. -10e- Pipturus argenteus (Forst.) Wedd. Urticaceae Group II or III (or I) Locality: Abaiang, no. 68 Pisonia grandis R. Br. Nyctaginaceae Group I Gilbertese: te buka Locality: Tarawa, no. 102 Tree not over 10 m tall; in other islands of the Pacific, it may be up to 20-25 m. Rare, found only as isolated individuals or small groups. Like many other species with useful wood, this tree must have been much more come- mon formerly, it has been much cut because of its soft wood, and replaced by the coconut palm. There have been various reports on the toxicity of soils under Pisonia trees, which would have been fatal to coconuts planted in their place. We could not find any justification of this idea and believe that the leaves and branches form a good humus. Uses: The soft wood is useful only to make outrigger floats. Pluchea indica (L.) Cass. Compositae Group III Locality: Butaritari, no. 61 Seen only/Butaritari, probably introduced from Ocean Island. Pluchea odorata Cass. Compositae Group III Locality: Bikenibeu, Tarawa, no. 22 Said to have been introduced by American troops who were said to call it "curray plant." Shrub up to 2-3 m tall, with mauve flowers and very fragrant leaves. Pluchea is one of the food plants of Chrysodeixis chalcites (Esper. ) Plumiera rubra L. Apocynaceae Group III Gilbertese: te meria English: frangipani Locality: Tarawa, no. 148 iy: Shrub about 2 m tall, with very ornamental flowers. Uses: The flowers are used in garlands. Polypodium scolopendria Burm. f. Polypodiaceae Group I Gilbertese: te keang Locality: Bikenibeu, Tarawa, no. 19 =103- Rather common but only in the shade, often at the foot of large trees. Uses: Leaves are used for the best garlands. The stems to perfume coconut oil used as cosmetic. Polyscias fruticosa (L.) Forst. Araliaceae Group III Locality: Abaiang, no. 77 Plant about 1 m tall, used for hedges by Europeans. Polyscias guilfoylei (Cogn. & March.) L. H. Bailey Araliaceae Group III Gilbertese: te toana Locality: Tarawa, Bairiki, no. 45; Teaoraereke, no. 5b Shrub up to 3 m tall, found only in villages. Uses: The fragrant leaves are used in garlands. Portulaca lutea Sol. ? Portulacaceae Group I Gilbertese; te boi English: purslane Locality: Bikenibeu, Tarawa, no. 18 Low fleshy plant, creeping or erect, much branched. Very common. P. oleracea may also be present, though not represented by a specimen. Uses: The leaves are eaten in times of scarcity. The plant is fed to pigs, and chickens are very fond of the seeds. Portulaca samoensis v. Poelln. Portulacaceae - Group I Gilbertese: te mtea. English: purslane Locality: Bikenibeu, Tarawa, no. 14 Low fleshy plant, almost creeping, with tuberous root. Uses: The leaves are boiled in several waters, and when soft, mixed with karewe and eaten in times of scarcity. The tuberous root is also cooked and eaten. The plant is fed to pigs. Premna obtusifolia R. Br. Verbenaceae Group I Gilbertese;: te ango Locality: Tarawa, Mission, no. 40a; Tearinibai, no. 40c; Butaritari, no. 40b Not very common, but must have been more abundant formerly. Tree 4-6 m tall, sometimes up to 15 m. -104- Uses: The wood is used in building houses. The roots perfume coconut oil for cosmetics. Straight sprouts make fishing poles. Formerly the wood was much appreciated to make fire by friction. Pseuderanthemum carruthersii var. carruthersii Acanthaceae Group III Gilbertese: te iraro Locality: Abemama, no. 7 Shrub up to 2m tall, seen any in villages. May have been introduced from the Marshalls. Uses: Flowers are used in garlands. The supple twigs are much ap- preciated to make fishing rods for ninimai (Gerres sp.) and other small fish: Pseuderanthemum carruthersii var. atropurpureum (Bull) Fosb. Acanthaceae 5 Group III Gilbertese: te iaro Locality: Marakei, no. 73 Shrub up to 2 m tall, purple leaves. Seen on all islands visited, cultivated as ornamental in front of houses, also in gardens of missions. May have been introduced from the Marshall Islands. Pseuderanthemum laxiflorum (Gray) Hubb. Acanthaceae Group III Locality: Tarawa, no. 110 Seen only in the garden of the Residency at Bairiki. Shrub 1 m tall, leaves green or green-purplish, flowers slightly mauve. Psilotum nudum (L.) Griseb. Psilotaceae Group I Gilbertese: te kimarawa Locality: Tearinibai, Tarawa, no. 58 seen only once at the foot of coconut trees in the Boariki area of Tarawa. Plant 25 cm tall. Uses: The extremities of the green branches, bearing yellow sporangia are used in garlands. Rhizophora mucronata Lam. Rhizophoraceae Group I Gilbertese: te tongo. English: mangrove Locality: Bikenibeu, Tarawa, no. 7; Aranuka, no. 93 One of the main components of the mangrove, usually found on the lagoon side and along the channels between islets; exceptionally found on the ocean -105- side, as on Bairiki, Tarawa, where 8 Rhizophora grow’on the windward reef. Their bases are submerged at high tide, but in spite of the breaking waves during ebb and flow the plants are able to persist. Rhizophora usually forms very dense stands, of average height of 4 m, but under exceptional circumstances, much taller. Thus on Aranuka, this species covers such a large area that it creates an illusion of two hills, even close-up. The trees grow in deep and foul-smelling mud, and may be up to 18 m tall and over 1 m around at the base. This stand, possibly because of its height, has been erroneously identified as Casuarina on charts of Aranuka. Uses: The dense and extremely hard wood is used in houses for rafters and sometimes for small posts. It is used also in stakes for fish traps, because it resists sea-water and ship worm (Teredo navalis). The bark pro- duces a red dye, rich in tanin, but is not used to color cotton nets. The bark is used to perfume coconut oil (cosmetic). Rhoeo discolor (L'Her.) Hance Commelinaceae Group III Locality: Abaiang, no. 156 Plant 30-40 cm tall, flowers white. Very rare, seen only on sites of ancient graveyards, especially on Abaiang. Ricinus communis L. Euphorbiaceae Group III English: Castor-oil plant Locality: Betio, Tarawa, no. 114 Not very common, seen here and there in villages, as ornamental. In Butaritari it seemed to be growing spontaneously away from the village. All the plants are well developed, sometimes over 3 m tall. The capsules are pink or sometimes almost red. The natives do not use them, and seem to be unaware of the purgative properties of the seeds. Ruppia maritima L. Potamogetonaceae Group I Gilbertese: te bukare Locality: Nikunau, no. 125 Small (40 cm tall) aquatic plant with white flowers, living completely submerged in the Nikunau lakes (fish ponds). Russelia equisetiformis Schlecht. & Cham. Scrophulariaceae Group III Gilbertese: te kaibaum Locality: Bikenibeu, Tarawa, no. 2 Plant 1 to 1.2 m tall, with scarlet flowers, found only in villages around houses. Uses: Flowers used in garlands. -106- Saccharum officinarum L. Gramineae Group III Locality: Tarawa, no. 141 Only two plants were seen, one at the Bairiki School, one at the mis- sion school on Butaritari. Both were poorly developed, not over 90. cin tall. Scaevola sericea Vahl Goodeniaceae Group I Gilbertese: te mao cans : Locality: Bikenibeu, Tarawa, no. 5 Most abundant shrub in the Gilberts, found everywhere: along windward beaches, across the whole width of the islands if the coconut trees are sparsely distributed, or absent, and on the lagoon side (leeward). Height from 2 to 4 m, stems * 30-40 cm around at the base, and branching from the base. Flowers white or yellowish; plant fruiting the year round. Uses: The flowers are abundantly used in garlands. It is a very common sight in the morning to observe young girls and children picking off flowers, or holding Guettarda leaves in which they are collected and taken to the village. The fruits, besides some uses in magic are said to be part of certain medicines. The branches are sometimes used for roof- ing strips, they are said to be less attacked by insects than those of other plants. Because of its great abundance, Scaevola plays an important role in protecting the soil against intense evaporation, and the leaves form humus. Sesuvium portulacastrum var. griseum Deg. & Fosb. Aizoaceae Group I Gilbertese: te boi English: sea-side purslane Locality: Nikunau, no. 126 Prostrate plant, about 30 cm, with white-greenish flowers, growing by the lakes. It is curious that it is not eaten by the Gilbertese who use Portulaca (also called te boi). Sida fallax Walp. Malvaceae Group I Gilbertese: te kaura Locality: Bikenibeu, Tarawa, no. 10 Low shrub very common all over the islands, especially on open or burned areas, where it may get very dense and tall (1.20 m). Uses: The flowers are used in garlands. The leaves are very im- portant in fertilizing babai plants (see pp. 87-88). Sida is the preferred food plant for larvae of Hypolimnas bolina, which are sometimes very numerous on it, especially in the sun. =1072 Solanum torvum Sw. Solanaceae Group III Locality: Bairiki, Tarawa, no. 109 Plant 2 m tall, with white flowers, seen only on Tarawa and Abemama. Sophora tomentosa L. Leguminoseae Group III Locality: Abemama, no. 91 Shrub said to have been brought from Ocean I. Only one plant was seen in front of a native house on Abemama; plant 2.5 m tall, non-woody shoots 1.80 m. No flowers or fruit. Stachytarpheta jamaicensis Vahl Verbenaceae Group III Gilbertese: te uti. English: blue cat's tail Locality: Bairiki, Tarawa, no. 104 Blue-flowered plant, not over 90 cm tall, very rare, seen only as ornamental around houses. Uses: The flowers are used in garlands. Stenotavhrum micranthum (Desv.) Hubb. Gramineae Group II Locality: Bikenibeu, Tarawa, nos. 29, 31 Very common grass, everywhere except in a zone about 20 m wide along lagoon and ocean side. Reddish in open areas, 25 cm tall. Synedrella nodiflora (L.) Gaertn. Compositae Group III. Locality: Butaritari, no. 74 Plant still rather rare, probably introduced by American troops. Seen only in Butaritari. Tacca leontopetaloides (L.) 0. Ktze. Taccaceae Group III Gilbertese; te makamaka Locality: Tearinibai, Tarawa, no. 135 Seen only once on Tarawa near the village of Tearinibai. The plants formed a dense stand over a rather large area under sparse coconut trees. Average height 1.60 m. Tacca is said to have been brought from the Marshalls about 15 years before by a Gilbertese who married a Marshallese woman. Some natives thought the plant to be poisonous, although no one could recall a case of poisoning. Perhaps their lack of interest in propagating this plant is due to the amount of work needed to make flour from the -108- tubers. This is regrettable, since Tacca could be a valuable resource as supplementary food. Its propagation raises no problem; its requirements are met by any of the vetter soils in the Gilberts and its resistance to drought is remarkable. The part above ground may wither entirely but the tubers resist and send up new stems as soon as the rain starts again. When the plants are pulled out to gather the tubers, a few always remain in the ground and give new plants. In soil of good quality at Tearinibai, Tacea showed very satisfactory development. The soil (fresh) had a pH of his tested with phenol red). The rhizomes were at an average depth of 9 inches. The total weight of all the tubers of | plants were: 2 lbs., 2.6, 1.9, 3.3. The largest tubers on one plant: & oz., 7, 4, 5, 4.5. The tubers are gathered when the flower stalk is turning yellow. They are washed in salt water and the outer skin scraped off. The tubers are grated with a stick tightly bound with coir rope and the grated pulp tied in a cloth and placed in a basin of sea water. The starch is forced out by pressing the cloth, and left to settle for 3 hours. Then the sea water is replaced and the starch settled again for 3 hours. . Finally fresh well- water is used and as soon as the starch has settled, it is spread out to dry in the sun. Because it is not extensively grown, the uses of Tacca are not well known. Still the owners of the plants ate the flour cooked in a porridge with kamaimai (toddy molasses); it can also be mixed with fresh toddy into a very thick pancake mix which can be fried. Buatoro is made up of Tacca flour, grated babai and kamaimai. Finally Tacca is used to make fritters called te tonati. The flour is mixed with sugar and baking powder, or, if the latter is unavailable, with fermented toddy. Small amounts of the very thick paste are deep fried in coconut oil or lard. Strangely enough, eggs are never used in native cooking, although a number of people have had many opportunities to learn to do so in European recipes. The stems. are said to be used in making hats but we did not learn the process for separating the fibers. Tamarindus indica L. Leguminosae Group III English: tamarind Locality: Tarawa, no. 134 Seen only on Betio islet, as 8 shrubs 3-4 m tall, normally developed. They were growing near a house and in a region where coconut palms and other plants are more luxuriant than anywhere else, so one cannot guess how well they would do in other parts of the archipelago. Tecoma stans L. cm In Mason's nursery the Leucaena plants probably sown in March and receiving various fertilizers reached sometimes larger sizes; some individuals mea- sured on August 20 had stems 35 cm long and roots 22.5 ecm. As pointed out in the chapter on soils, a careful examination of the roots showed no nodules. Trying to inoculate the roots with nitrifying bacteria might improve the development of these legumes, but this would be worth while only if not limited to a small area. To conclude, it does not seem that even Leucaena has much future in the Gilberts. To be really useful as a soil improver, such a legume should be able to grow without care, once sown, and to cover large areas. One of our nurseries, where the seeds were sown very close together might have led one to think that the plants could be used as green manure, but the price of seed gathering in Fiji or elsewhere would make the project too expensive. As for the seeds planted about 1 m apart, in unprepared soil, they all came up but their growth was even slower than that of the others. On an area of about 1/2 hectare, children to whom we gave bags of such seeds were happy to sow them under these conditions, and it would be interesting to know what will remain a year from now of these thousands of seedlings. 22436 CHAPTER 10 INSECTS COLLECTED IN THE GILBERTS During our field surveys, particular attention was paid to the collec- tion of insects most frequently found on economic plants, but time was ob- viously lacking to organize methodical collecting expeditions. Our col- league, Mr. R. Mason, Agriculture Department, Suva, showed commendable initiative in making general collections which would have represented a valuable contribution to the study of the entomological fauna of the Group. We were sorry to learn recently, that these collections had been almost entirely damaged in the ship's hold. Several of the insects mentioned may be considered as more or less dangerous pests, others are merely present without being a threat. Some are new in the Gilbert Islands. It is obvious that a methodical investi- gation of the entomological fauna of these islands would appreciably add to the present knowledge, from the biological and systematic point of view. A detailed study of some groups such as Hymenoptera, for instance, would certainly yield valuable data in connection with the biological balance observed. It seems, indeed, that the small extent of the damage caused by insect pests may be imputed to the large number of their predators. The following list, and the enumeration of specimens still being identified will represent only a minute contribution. In order to give a more comprehensive view, it would have been essential to pursue this single aim only, and a much more considerable period of time would have been necessary. The polyphagous nature of some species, such as the caterpillar of Prodenia litura (F.) (very harmful) and that of Hypolimas bolina subsp. rarik Esch. must be mentioned. Their adaptation to recently introduced plants is sometimes extremely rapid, as we observed when we found cater- pillars of Hypolimnas on Phaseolus calcaratus in our Bikenibeu nurseries, six weeks after sowing. List of Insects Identified LEPIDOPTERA Achaea, janata (Linné) -- Family Phalaenidae Caterpillar on Pemphis acidula Forst. Agrius convolvuli subsp. distans Butler - Sphingidae Caterpillar on Ipomea tuba Amyna octo (Guénée) - Phalaenidae Caterpillar on Abutilon indicum (L.) Sweet Anisodes ceramis (Meyrick) - Geometridae Host unknown -114- Anua coronata (Fabricius) - Phalaenidae Caterpillar on Terminalia catappa L. Badamia exclamationis (Fabricius) - Hesperidae Caterpillar on Terminalia catappa L. Cephonodes picus Cramer - Sphingidae Caterpillar on Guettarda speciosa L. Chloauges woodfordii Butler - Pyraustidae Caterpillar on Guettarda speciosa L. Chromis erotus (Cramer) - Sphingidae Caterpillar on Morinda citrifolia L. Chrysodeixis chalcites (Esper.) - Phalaenidae. Caterpillar on Pluchea odorata Cass. and Solanum lycopersicum L. Danais plexippus (L.) - Nymphalidae Caterpillar on Asclepias curassavica L. Decadarchis sp. - Lyonetiidae Caterpillar on Cocos nucifera L. Ethmia colonella Walsingham - Ethmiidae Caterpillar on Cordia subcordata Lam. and Amaranthus dubius Mart. Hypolimnas bolina subsp. rarik (Eschscholtz) - Nymphalidae Polyphagous caterpillar. Preferred food plant: Sida fallax Walp. Nagia hieratica Hampson - Phalaenidae Host unknown Precis villida villida (Fabricius) - Nymphalidae Caterpillar on Scaevola sericea Vahl Prodenia litura (Fabricius) - Phalaenidae Polyphagous Spodoptera mauritia (Boisduval) - Phalaenidae Host unknown Utetheisa lotrix (Craner) - Arctiidae Caterpillar on Messerschmidia argentea (L.f.) Johnst. COLEOPTERA Conoderus pallipes Eschscholtz - Elateridae The larva attacks those of other insects. ass Cylas formicarius (Fabricius) - Curculionidae Considered as very harmful to sweet potatoes, but only found on Ipomoea tuba. Diocalandra frumenti (Fabricius) - Rhyncophorinae Found on many coconut palms but never in considerable numbers. None of these trees gave signs of injury which could be attri- buted to damage by this insect. Harmonia arcuata (Fabricius) - Coccinellidae This predator is very common everywhere. Lacon modestus (Boisduval) - Elateridae Necrobia rufipes (de Geer) - Cleridae This species of Cleridae is found in considerable numbers chiefly in copra lots. Papuana sp. or Pentodon sp. - Scarabaeidae Pest of Cyrtosperma, Musa and possibly of young coconut palms. See p. 26. Sessinia livida (Fabricius) - Oedemeridae This insect is called "“toddy beetle" and is found in great numbers. It has very bad fame in Gilbertese circles, and any toddy container in which one of these beetles is found is immediately emptied. Grimble writes that in the past, these toddy beetles were used for criminal purposes; the designated victim was given a drink mixed with the juice of a definite number of these beetles. The effects on the bladder were said to be frightful. ORTHOFTERA Cutilia soror (Brunner) - Blattidae Graeffea cocophaga (New.) - Phasmidae Pest of the coconut palm (see Chapter 4, p. 28). DERMAFTERA Chelisoches morio (Fabricius) HETEROPTERA Pachybrachius nigriceps (Dallas) - Lygaeidae DIPTERA Chrysonya megacephala (Fabricius) - Calliphoridae Dacus (Strumeta) frauenfeldi Schiner - Tephritidae ~116- Homoneura acrostichalis (Meijere) - Lauxaniidae Pseudeuxesta prima Osten Sacken - Otitidae Rhinia testacea R. Desvoidy - Calliphoridae Sarcophaga dux Thomson - Calliphoridae HYMENOPTERA Megachile diligens hedleyi Rainbow - Megachilidae Megachile fullawayi Cockerell In the Gilbert Islands, the Megachiles do not affect palms, but prefer many bush plants, particularly Morinda citrifolia L. ‘They cut circular pieces in the leaves for use in their nests. Odontomachus haematoda (Linné) - Formicidae Extremely painful venom. Pachodynerus nasidens (Latreille) - Vespidae Pheidole umbonata Mayr - Formicidae Polistes fuscatus aurifer Saussure - Vespidae Szepligetella sericea (Cameron) - Evaniidae Alaife CHAPTER 11 MARINE RESOURCES PART I: GENERAL SURVEY The value of the resources drawn by the natives from the sea is con- siderable. It is important enough during normal periods, when plant prod- ucts are found in abundance, and even more so when these resources fail, exhausted by prolonged droughts. Thus in the southern islands, fish, crustaceans, and molluscs are a substantial help for populations tempo- rarily affected by real shortages of food. Sea foods obviously do not have a dietary value equivalent to that of products such as grated coconut and especially toddy, which is so essential, but they bring a quantitative contribution with which the native is able to hold out until the return of normal conditions. It is often believed, partly because of the extensive activity form- erly associated with deep sea fishing such as fishing for Ruvettus that the natives chiefly fish in the open ocean. Of course some fishes can usually be caught only at considerable depths - 100 fathoms and over - and for others canoes must go several miles off-shore, as in bonito fishing. Around some northern islands the capture of flying fish and tuna involves rather elaborate means and great qualities of seamanship. But these spectacular operations are not everywhere a daily occupation. After observing the various fishing activities of these territories as best we could, we are now convinced that apart from these particular regional cases the fish supply of most Gilbertese populations is found much more frequently in the immediate vicinity of the land and in the shallows than far offshore. This observation is even more true of atoll islands, since the lagoon and especially its shallow parts are the most commonly frequented fishing grounds. It is also the place where a maxi- mum subsistence may be found with a minimum of effort. In some privileged islands, the sea brings to the door, so to speak, fish or other edible organisms, sometimes in considerable quantities. This is the case with ikari, a fish of the genus Albula, and of baitari, edible jelly fish of genus Tamoya (family Carybdeidae). The native utilizes almost all the varieties of fish he is able to catch, provided they are not poisonous, and very few are. (It should, however, be noted that some fishes which are commonly eaten in some islands are considered as toxic in others.) While the native very seldom eats chicken or eggs, he will not scorn even such species as diodons and tetradons, in which there is really very little to eat once the toxic parts have been re- moved. It is an interesting fact that cases of poisoning by shell-fish or fish are extremely rare. PART II: DEEP-SEA FISHING Formerly, the Gilbertese went fishing in groups for Ruvettus preciosus Cocco, using remarkable techniques and going far out to sea, out of sight of land. This fishing is no longer practiced on a wide scale, both because the Governnent has forbidden canoe trips from island -118- to island (except between Tarawa and Abaiang) and because the natives of today have completely lost the remarkable science of long-range navigation and astronomical reckoning which their ancestors had developed. A few Ruvettus are caught not very far from shore, but they are not sought in preference to other varieties. The following species are still sought away from the shores of the islands, although never out of sight of land: large tuna, (Neothunnus macropterus Sch., te baibo), bonito (Pelamys sp. te ati), and small tuna (Neothunnus Sp, 5 Le natiati)(l . Their presence is indicated, as every- where else, by hundreds and often thousands of sea-birds coming to join the predatory fish in their feasting on shoals of sardines and flying fish. These great flocks of birds, which present a most lively and noisy spectacle, are always a promise of productive fishing for the natives. From far away they can assess the size and behaviour of a shoal of bonito or tuna by the appearance of the birds. When the birds are densely con- centrated, flying very near the surface, diving quickly and often, the prespects are excellent, while dispersion, hesitating flight at a greater height, and few dives will indicate a scattered and not very important shoal. But whether the signs are more or less favourable, the alarm is given as soon as they have been noted, and the canoes leave. All the better if the wind is in the right quarter; they will be on the spot sooner and with less noise. Sometimes six, seven or eight canoes arrive together at the seething swirl caused by the preying fish. The method used is a kind of trolling. The equipment consists of a line 25-30 m long; it may be shorter or longer according to circumstances. This is followed by a wire leader on which a feather lure is mounted, made of cock feathers predominantly red, tied on to a fish-shaped wooden body and armed with a single hook, very sharp and barbed. The Gilbertese seem to disregard completely the use of the spoon but are not surprised to find that it also catches many fish. Actually metallic lures obtain such good results only because they are used with nylon lines which ob- viously require a reel with adjustable drag. Such a method being only attractive as sport, there is no reason why the natives should take an interest in it. Cruising to and fro through the shoal and along it, the fishermen often have barely time to haul in and unhook the fish, and pay the line out again until the fish scatter. It is not uncommon for each canoe to bring back a dozen fish weighing altogether around 130 pounds. These expeditions are all the more productive when a moderate breeze enables the canoes to sail regularly and noiselessly without using the paddles, which disturb the foraging fish. (1) Te natiati is the name given in Marakei to the species called te baiura in Tarawa. “1198 Fishing for Great Tuna (te baibo: Neothunnus macropterus) and Some Other Predatory Fishes Different fishing practices are used for the red-fleshed yellow-fin tuna, and it seems that it is chiefly in the north and centre of the Gilbert Islands that this fish gives rise to a frequent if not regular activity. It is productive chiefly from April to September, but isolated specimens are found all the year around (in the south the best tuna fish- ing occurs from October to March). It is no longer the chance appearance of bird concentrations which tells the Gilbertese that it is time to fish; they rely rather on their great experience of tuna migrations at fixed seasons, in relation to some factor or clue the knowledge of which is handed down from one generation to another. The methods used will be either trolling in the daytime at variable distances, generally over two miles from the shore, or drift fishing without sail by day or night at similar distances. Finally, the natives fish for te awai (Aprion sp.) chiefly at the entrance of the lagoon passes where they also catch the small tuna and bonito already mentioned. Drift fishing: The great tuna are also caught by drift fishing using weighted lines baited with flying fish (te onauti = Cypselurus aD. )¢a 4 BY day, the bait will be let down to a depth of 30 to 50 fathoms (in some southern islands such as Nikunau, baibo measuring 0.30 to 1.80 m are caught at this depth). By night the tuna will bite at between 50 to 70 fathoms and some times te ikabauea (Sphyraena) will be taken at the same depth. If the tuna do not bite, the fishermen go a little farther out at sea and may hook te ikanibeka (Ruvettus preciosus) at depths varying from 90 to 160 fathoms with a special hook which has often been described. It seems that this "castor oil fish" was much more highly appreciated in the past, but we were unable to find the reason why. The natives laugh every time ikanibeka comes into the conversation since the flesh of this fish, unless cooked several times, has purgative results of ir- resistible rapidity (as indicated by its name). It is obvious that we have mentioned here only the fishes especially sought by these methods, but others are often caught in addition, ubic- uvitous species which chance and circumstances bring as a supplement, such catches will be all the more frequent when the natives use lines with several hooks at various depths, as is often done in the north, especially around Abaiang. Sharks are also caught in these regions by drift fishing at depths of from 20 to 30 fathoms, and often with the help of large shells (te ang). Trolling (te katiki): This method derives its chief advantages from the speed and remarkable manageability of the outrigger canoe, and it is practised with the equipment already described for drift fishing. The only difference is that the hook and sometimes the line will be stronger. -120- The latter is fixed to the canoe with or without a slip knot according to the area where the great tuna are caught. Their average weight seldom exceeds 40 kilos, and it seems that specimens of 20 to 25 kilos are most common. We did not stay long enough in some islands to form an exact idea of the number of tuna which may be caught, but,we were told that the natives voluntarily limit themselves to the quantity necessary for their family or, to what they are sure of selling in their village. In the event of "overproduction" the excess would eventually be processed but the Gilbertese do not seem to like processing tuna, which is too oily and dif- ficult to keep in good condition. Sometimes, while a canoe is engaged in trolling, swordfish may strike the lure and cause the fisherman much worry, especially if they are large. Some such occasinns were described to us when rakuriri (Istiophorus) or rakuika (Tetrapturus) fought for a long time without the fisherman being able to do anything except let his canoe be towed or wait until the fish was exhausted to land it. On other occasions these swordfish were so large that there was no hope of subduing them or of towing them to the island, and the only thing to do was to cut the line. Finally, we were told of two fatal accidents, the details being confirmed from several sources. One happened a few years ago and the other more recently. On both occasions a large swordfish was hooked, and after fighting for a while made straight for the canoe and impaled the unfortunate fisherman. Other fish also caught by fast trolling are te ikabauea (Sphyraena sp.) a species of barracuda rather common in these waters, and te baara (Scouberomorus sp.) Flying Fish Catching flying fish occasions great activity at certain times of the year, the seasons varying widely from the north to the south, They are used partly for bait for tuna and partly for food. These flying fish (Cypselurus sp. = te onauti) are found in greatest abundance between April and September in the northern and central islands, and from October to March in the southern group. They are never caught in the lagoon, but outside the breakers on the ocean side. They are noticeably more abundant around lagoonless islands (reef islands) and also at Marakei (which, though having a lagoon, offers an outside shore similar to those of reef islands). A distinction is generally made between three main fishing methods: (i) Daytime fishing with special trolling lines equipped with floats (te ai-onauti). We have not observed this and ob- tained this information from I. G. Turbott. (ii) Twilight fishing, which begins a little before sunset, lasting till darkness is complete (te kababa). -l2l- (iii) Torch fishing at night with dip nets at the time of the full moon (te tatae). The natives are said to state that the best night is that of the full moon, and that if the fish are abundant the three nights before the full moon and the three following, are also good. These indications, which were con- firmed by Turbott, may be compared with those concerning massive concentrations of te ikari (Albula vulpes) along the shores outside Tarawa Atoll (see p. 132). We had the good fortune in March to take part in a torch-fishing ex- pedition off the shores of Marakei. Of this most striking and beautiful spectacle, with flaring torches illuminating sea and canoes, only the technical aspect will be described; ethnological details such as rites and customs associated with this activity will be omitted. At the bow of each of the dozen or so canoes each paddled by 2 or 3 people, including some women, stood a man holding in his left hand a _coconut leaf torch, in the right a large, long-shafted dip net with which he caught with the skill of a butterfly catcher, the fish attracted by the light. For one canoe we counted one flying fish caught approximately every three minutes. It is not uncommon for the catch to be larger, and it may include over 100 fish after a few hours of careful fishing. The least successful expedition results in a dozen fish per canoe. The system of distributing the fish caught at night differs accord- ing to the islands and to the particular agreements which the canoe owners conclude with fishermen, paddlers and steersmen. In the northern islands it is customary when fishing has been productive to reserve a large portion of the catch for preservation by sun-drying and salting. The rest is eaten fresh, except of course the fish intended for tuna bait. The delicate meat of Onauti is appreciated by natives as well as by Europeans. It is said that the Gilbertese like this fish for the oil contained in its head. PART III: REEF AND LAGOON FISHING Fishing on the reef on the ocean side The natives actually "glean" rather than “fish" whatever they can find which is edible. Generally comprising small but widely varied spe- cies, the catch, after a few hours' gathering, represents a substantial addition to the routine family meal. We often observed with amusement - and this is a typical sight in the Gilberts - the similarity between the natives and the herons, both hunting on the same reef at the same time for the same prey, most often a few yards from each other. Criss-crossing these large flats in all directions, they give the same methodical care to their search. The spear and the bill do the same work; each crack, each hole is explored. The native obviously has the advantage of being able to turn over heavy stones and thus uncovering rich puddles. The heron will very soon follow and complete the inspection. -122- Which species are iiost commonly seen in the basket of a fisherman returning from his almost daily tour? A few crabs, small morays, fishes of which the largest do not exceed a hand's width, shell-fish and almost always small octopus. To catch the moray eels hiding under rock edges, the natives use an umbrella rib or some other metallic spike curved at the end. When the eel is forced out after much prodding this flexible spike will be used to whip it until it is easily grabbed. For added safety the natives will sometimes bite these eels behind the head to kill then. Alternatively a noose at the end of the spear is used. It is at night that the most extensive activity can be observed on the reefs. Coconut leaf torches or kerosene lamps are used. In every puddle some fish trapped by the receding tide is blinded by the intense light and easily captured. The fishermen, however, seek mostly the young octopus. New moon nights are the best. The small octopus is caught by hand at the moment when it flattens itself on the sand to hide. Here again the natives will kill by biting. These small specimens of octopus (kikao) are preferred to any others both for food and for bait. As every- where else they are one of the best bait obtainable. The large specimens of octopus (kika) are caught only during spring tides on the reef enclos- ing the lagoon. This fishing is done at the new or full moon, and also by © day. fuk Shellfish: On the whole most of the edible mollusks gathered on the lagoon side are bivalves. The shells of other groups are gathered on the ocean side reefs. They are in order of preference: Turbo setosus Giel. ; (te nimatanin) Circe pectinata Linné (te koumara) Tellen scobinata Linné (one of the species called te af ¥ koikoi) Drupa ricinus Linné (te kerekaka) Modiola agripete auriculata Krauss ee nikarinei and te katati) Strombus gibberulus Linné te newenewe ) Atrina nigra Dillwyn (te katete) Conus lividus Hwass (te anikomri) Gathering shellfish is the privilege of the women and children and is done by scratching in the sand of the uncovered flats at low tide, both on the lagoon and ocean sides. Koikoi (Asaphis deflorata) are found marked by little holes and one or several may be dug out, with a piece of coral or coconut shell, at a depth of 10 to 20 cm. Another bivalve, smaller but of more delicate taste, will be gathered a few centimeters below the surface of the sand along an almost regular line in the wet zone just left by the ebbing tide. This is te katura (Atactodea glabrata Gmel.) which is eaten in great quantities. We have sometimes observed large baskets full of these shells being brought into centres such as missions, where the practice of collective meals is more common than elsewhere. In the vil- lages, each family usually gathers its own supply of katura and other mollusks. Te nouo (Strombus luhuanus) is also frequently eaten, especially in Abemama, where it is more common than in other areas. eile Finally particular reference should be made to the Tridacna (te verevere) of all sizes, and especially to Hippopus hippopus Lam. (te neitoro) which the Gilbertese like very much. Fishing for these giant clams is practised at the entrance to some lagoons in atolls such as Onotoa and Aranuka, where they are found in abundance in the living coral reefs. This activity is carried out by divers on te rakai (shal- lows) and requires the use of goggles and a strong knife. The diver quickly inserts the latter between the two valves of the shell which close and strongly grip the blade, making it easier for the diver to free the shell from the bottom. These large bivalves are generally eaten fresh, but may be preserved after sun-drying. This type of fishing is also practised at the edge of some outer reefs, weather - and sharks - permitting. Edible jellyfish: Another organism of the ocean side reef which is highly appreciated by Gilbertese natives is te baitari, jellyfish of the family Carybdeidae, genus Tamoya. Every month - seven days before the full imoon - the sea casts up on the shore, on the ocean side, large quantities of this jellyfish (Plate XVa). Only the inner part is edible. The natives prepare them on the spot, by scraping away with a knife all the outer gelatinous parts. The rest is spread on strings stretched between two coconut palms or two stakes, and sun-dried for two or three days accord- ing to the weather (Plate XVb). ‘The final product, which in appearance faintly resembles tripe or large sausage skins, is placed over pit-ovens and left on the very hot stones until crisp. It is then eaten, generally with babai, breadfruit or grated coconut. Baitari may also be cooked in water. We made a point of tastingit, cooking it like fried potatoes. Fried crisp, and well-drained, baitari is really delicious. While baitari is not a basic element of native diet, it is one of the little side-dishes which are an addition to the daily fare. One can form an idea of the pleasure they give to the Gilbertese by observing how eagerly they eat them. We may be able to learn their real food value from analyses which will be made later on. We made arrangements for a sufficient quantity to be collected during the last full moon of our stay, but unfortunately, a contrary wind blew during the whole of the preceding night, and no jelly-fish were thrown on shore. It was then too late to go back to the opposite side of the atoll, which benefited by this wind. We could only leave instructions with reliable persons that a shipment of dried baitari be made during the next month. Fishing at the edge of the reef This fishing technigue is very interesting and is practised both by men and women. Each fishes for himself, and 5 or 6 fishermen avoid getting in each other's way along a kilometer of shore line. At the ris- ing tide, the fishermen follow the zone where the rollers, after breaking violently, rush into the identations at the reef edge. The fishing gear includes, in addition to a little basket tied to the belt, a pole of the greatest possible length, a line made of fine string and sometimes weighted -12h.- with a small piece of coral, and a hook baited with a fragment of shell- fish or hermit crab, or even better, of young octopus. The bait follows the alternating movements of the continual swirl, so that the hook is very seldom snagged. The same species of fish, te reiati (Cirrhitus ) is almost always taken, seldom over 20 cm in length. A catch of 20 to 30 reiati taken in a few hours is considered quite good. This is the only fishing technique requiring less sketchy gear than usual, since the pole must not only be long but also flexible. A native addicted to this method tries even when his means are limited, to buy a tall bamboo pole (this plant does not exist in the area) for which he will not hesitate to pay the 8 or 10 shillings required at the Betio store (Tarawa). Although we have very seldom seen it done - quite understandably - we will also mention that young sportsmen even go and stand on isolated coral heads right in the breakers. Their chances will obviously be greater, but so will the efforts required to keep on their feet. One can- not but admire the agility with which they cut through the waves, or dis- appear under them without losing hold of their gear, only to start fishing again, taking advantage of a series of lesser waves to catch the fish which bite better at such a distance from shore. The species most regu- larly taken is te koinawa (Hepatus triostegus ). The experts sometimes take more than 20 in one hour. Fishing outside the reef This type of fishing is done in canoes, and must have been the most popular in the past. The canoes were more numerous » and so were the men able to use them and to make the most of this type of craft in often choppy waters. Now, only three or four canoes at most are seen along a reef several miles long, such as the Bairiki reef (Tarawa), and only in very good weather. They generally keep at distances affording depths of 15 to 20 meters. There are two methods. In the first, the canoe is anchored and the fisherman uses a long weighted hand-line. The hook is generally rather large and baited with a piece of fish or octopus and held on the bottom or a little above. Serranidae and Lethrinidae and generally sedentary species typical of rocky or coral ground, are taken in this way. With the second method, the canoe is allowed to drift with the cur- rent or with a slight breeze, this drift being checked now and then by a stroke of the paddle. The line is tied directly to the canoe, with a loop tied with a slip knot to serve as a visual indicator and to hook the fish all at once. The same fishes will be taken with this method as with (I) Although there may be less canoes nowadays at Tarawa this should not be the case on other islands. (Note by the Administration of the Gilbert and Ellice Islands Colony. ) -125- the preceding one, with the addition of a few species which prefer a mov- ing bait to a still one: Carangicae, Sphyraenidae, etc. These two techniques, used with a bottom or a half-depth line, are always productive. In four or five hours, it is not unusual to catch seven or eight fish with a total weight of from 10 to 20 kilos. A few fishermen will venture farther off shore, especially when they intend to catch sharks. The method used is drift fishing with "hanging sail". The gear includes a rope line followed by a chain and a sharp hook with small barb. Fish will be used as bait. The hook will be let down to 20 or 30 fathoms. It must be mentioned that some Gilbertese fishermen have the greatest faith in a "call" produced by knocking together several large shells tied to a rope and immersed at a slight depth. These helmet shells are Cypraecassis sp. probably C. rufa (te ang). The characteris- tic rattle (te kakerukeru) is said to attract sharks without fail. We were not able to try it. Shark fishing is not a frequent occupation. Rather it is considered as a sport. It is practised only in groups. The offers made by the co- operative officer for shark fins do not seem to incite the natives to go fishing more often. Actually one must admit that the fishermen are not equipped for large-scale operations; they prefer to fish for food than for uncertain long-term profit. Sharks are. cumbersome beasts which cannot easily be hauled into fragile craft, and have to be towed back to shore once they exceed a certain size. Some fishermen told us of letting go a lh meter shark in order to resume fishing for te kuau (Serranidae) or te rou (Lethrinidae). They even neglected to take the fins although knowing that co-operative organizations would buy then. Scarcity of fish at certain times Besides the natural factors which determine the abundance or scarcity of some species of fish at particular seasons or times, two occurrences may be discussed. The first is the scarcity of great (yellow-fin) tuna around Marakei Island at a time of the year when they are normally numerous. According to the natives, this scarcity was due solely to the quantity of rain which had fallen in the area for weeks. The fact itself was indisputable, but the explanation was less satisfactory. The great tuna comes near the shore under favourable circumstances. It is known that it can live only with high salinity and temperature, and seeks these conditions especially at spawning time. It is possible that during periods of heavy daily rains such as fell at this particular time, these conditions are not realized. A decrease, even temporary, in the surface salinity might have a direct influence on tuna, or it might effect them indirectly by having an influence on the shoals of flying fish which they follow, or even merely on the plankton on which flying fish feed. Only a special survey carried out on the spot could give a reliable answer. -126- The second occurrence is the alleged scarcity of fish in the southern group of islands, following intense and prolonged droughts. Grimble (1933- 34) says: "...in times of drought, when not only vegetable foods but all kinds of fish are scarce..." While the influence of rain might easily be admitted, this coincidence of drought and fish scarcity is difficult to accept. No definite confirmation was given by the natives of the terri- tories in question, but this belief is so firmly held in some circles that it may yet express an observed fact. A tentative explanation may be of- fered until more information is secured. Since the species involved are mostly fished for from canoes and outside the reef, the natives, during long periods when food procurement is difficult, would be less prepared to make the physical effort required for offshore trips, as a certain apathy results from food shortage during these hard times. Lagoon fishing Surprisingly few large fish live in the lagoons, which are often quite vast and sometimes deep. The only predatory fish commonly found in them are ikabauea (Sphyraena sp.) and tauman (Caranx sp.) both of rather small size, foraging among the shoals of small fish inhabiting in considerable numbers these slightly clouded waters. Aua (Mugilidae) and ninimai (Gerres sp.) predominate, and are each represented by several species. Another important member of this lagoon fauna is tarabuti, a clupeid of the genus Harengula, which is caught in abundance in the vi- cinity of channels, wharves, and generally in the deepest parts of the shallowest areas. Net fishing: Lagoon fishing is practised at high tide on the immersed reef flat. Purseless seines are used (a type of purse seine, te rienan urakaraka, was mentioned to us but we have never seen it), with mesh of variable dimensions according to the type of fish sought, sardine mesh, whiting mesh, and mullet mesh. For the latter the seines are made of thread as fine as possible so as to be less visible. The nets are made of cotton; coir seines have become extremely scarce, though cotton nets may be seen mounted on coir ropes. The floats are most often of te kanawa ‘(Cordia subcordata Lam.) wood, while weights are generally made of shells (te koikoi, Asaphis sp.). The nets with the largest mesh are used to catch the species called by the natives "angry fish" (ika-nun), for in- stance the Carangidae (te rereba) and to catch ikari (Albulidae). The medium-meshed nets are chiefly used to catch Mullidae, Gerridae, Lutjani- dae, etc., and the very small meshed nets only for Clupeidae (te tarabuti = Harengula sp.). We were also told that in Abaiang Island small nets with very close mesh are used for te maebo (Mullidae) and inos- quito netting for sardines (te tarabuti). Fishing by casting-net is very little practised in the Gilbert Is- lands. We have only seen these nets used once in Tarawa by a few fisher- men who did not display very great skill. They were fishing near a wharf where mullet, sardines, and whiting were concentrated. The ways of using the seines are of the greatest simplicity, but are always practised with great skill. In a place where fish shoals can be Lene detected, either by the characteristic shimmering caused by fish swimming near the surface or by the presence of a few sea-birds taking their toll of the small fish, a group of men, sometimes accompanied by a few women, advance slowly on the sunken flat. They move noiselessly in water up to their waists, dragging behind them the seine already spread out. When they are near a shoal of fish, they rapidly arrange the net in a circle. While two men carry out this maneuver, the others beat the water in the area facing the opening of the seine. Once the net is closed the fish rapidly enmesh themselves, although some always escape by leaping over the top. The natives do not regret their loss since so many more are caught. All that is left to do is to take them off the net. They are killed with a bite behind the head and thrown outside the net, where they are collected in small baskets. Then the whole operation is resumed a little further on. This method of fishing is generally productive, and without much effort full baskets have been carried ashore several times in a couple of hours. All the fish are then tied together with a coir string and hung from a pole carried on strong shoulders and brought to the village (Plate XVIb). These loads often exceed a hundred pounds and provide the fishermen not only with food for themselves and their families, but also with money, since a pound of fish may be sold on the spot at an average price of three pence. ' This fishing method mostly brings in Mugilidae and Gerridae of small size, but the dull color of a small shark (generally Mapolamia spallanzani ) is sometimes seen among the silvery glitter of the other fish, or, one oc- casionally observes the bright colours of some Canthidermis or Scarus caught near coral formations. At the edge of these lagoons - but only in the vicinity of channels between the islets where the water is more often renewed and somewhat less murky - there are small coral heads, not very luxuriant, around which are found, as in a miniature replica, the usual animals of the ocean side reef. The fishermen take no chance of snagging . their nets, but often work them near these coral growths. In addition to the extremely simple method described above, two other more specialized techniques are practised mostly in the north and especially in Abaiang. The first of these is used for capturing Albula (te ikari). The principle is illustrated by the sketch (Fig. 18). The fishermen place their seine in a given spot where they think the fish are likely to go. The helpers come up behind the shoal and try to drive it towards the net. If they succeed, they move faster and the fish are rapidly driven into the seine, which is then closed as quickly as possible. The second method, illustrated by sketch (Fig. 19), is designed solely for the capture of mullet (te ava) and differs from the first in that the seine, instead of being wide open, has both its ends very close together. Whenever a shoal of mullet appears in the vicinity of this opening the presence of the net and of the two men holding it frightens them. They swim off. At this moment the helpers scare them suddenly and they turn around. The mullet rush into the net, which is immediately closed and can then be lifted since most of the fish have enmeshed themselves at once. The success of this method is due to the fact that mullet are very easily frightened and very swift-moving. La OBe Method of net.fishing for ‘i KANUN and “IKARL ‘ Fig 18 The wide-open net is placed in front of the ¢§f Fish which vane ‘driven, into iE by a beater. eS of net -fishing for \\ wl AUA Fig. AQ The half- open net is placed behind the fish which rush into if in endeavouring to escape the beater. Bi Fe wo types of fish traps in the Gilbert Tslands Sree oe oe ce Se B fs | (0 os ae pee xe SS = - TT no — \S Baa hla ee en ‘ ~ eS “Im a Bb gee te # a ae : ‘ ‘ a = =I ee ie Neg ak ie Se aes ' ¢ ott eS 2 aes ties ae Bib ae g eee a 217) ==) ae LOE = aS) vat Ly ‘s ia ae Z Ss, > a ie eres > ue puigne ya Hy 1 a Vy « an ay In fact this "specialist" will have no other task than that of observing the growth of chanos and advising when they have reached a size justify- ing their capture. They are then 70 or 80 cm long. Although chanos fry (te tawaa) may be found at any time on the shore of the island, it is gathered only twice a year during spring tides. The tiny chanos are caught at low tide in the reef puddles with nets of very small mesh made of mosquito netting or even canvas. The east-south-east area of Nikunau is said to be the richest. The lakes are stocked as the fry are caught. They are transported in all kinds of containers, even in coconut shells. The adult chanos in the lakes are fished only once a year. Sometimes the natives wait for two years if the fish have not reached a sufficient size. All the inhabitants of the island have a right to participate. This fish- ing is done with moving nets, which are more efficient on account of their number and mobility than of their dimensions or shape. They are usually simple large-mesh nets mounted on two wooden stakes similar to those placed at the ends of a seine. Except for the few chanos which manage to escape, the whole reserve is fished out. The fish which is not eaten fresh is processed (salted and sun-dried). Some will be kept for a time in the huts, and some will be sold at the co-operative centre (Boboti). The Nikunau and Beru natives spoke highly of the remarkable quality of the baneawa, which is considered superior to all other fish. When asked why they were not content with catching adult baneawa on their shores, they specified immediately that the baneawa coming from nei (fish ponds) had much more fat on their backs. In this area, which is ideal for the pond culture of chanos, we did not see a single specimen of baneawa. In other regions we have, on several occasions, examined the contents of the stomach and intestines of this fish and always found a mixture of mud and algae (bukabuka). PART VII: MISCELLANEOUS MARINE RESOURCES Trepang (Holothurians ) Lacking adequate equipment to make the necessary marine surveys our- selves, we enquired from the natives about the possibilities of catching marketable holothurians. But no one was ever able to bring us reliable information or interesting specimens. Those brought to us were always of the common and valueless species which we saw for ourselves on the un- covered reef-flats; te ntabanibani, te uniganikakua, te nei karua kere- boki, the prevailing species was most often te riburibunimainiku, of the Aspirochirota group. We were told that the teat fish, a holothurian of the greatest value (Holothuria mammifera) exists in this area, but this information needs to be borne out by specimens. This sea cucumber is said to be called in the vernacular te uningauninga. However, even if the con- mercial species were present in sufficient quantities in some areas of the archipelago and at depths where fishing would be possible for average divers, the Gilbertese would feel no urge to undertake such work. It seems, in fact, that trepang fishing has been abandoned for two or three genera- tions. -139- Mother of Pearl Shells We found the same lack of information in relation to trochus shell, of which we never saw a single specimen. As to pearl shells, they are said to be found around Abaiang and Onotoa, and seem to be called te paio(2) in the vernacular. It appears that there is some confusion in the minds of some natives who also gave us the name te katai. But this latter word actually applied to Atrina nigra. Sponges In spite of the absence of trade in sponges(2), they are found in abundance in some areas and particularly in the Marakei lagoon where nui- erous specimens were gathered for us belonging to the genus Euspongia, probably E. irregularis. The sponges found on the lee side of some la- goonless islands such as Nikunau undoubtedly belong to this species (vernacular name te ongantari). Substantial samples were sent to various firms in Great Britain. We did not see the documents estimating their worth, but it appears from verbal information that these sponges are very inferior to those of the Mediterranean and could find only an industrial use in filters. In any case, it appears to us that the stock is quite in- sufficient to supply a market. Shark Fins Sharks are plentiful around all these islands but, as mentioned earlier, they are fished for only occasionally. Just after the war, the Co-operative Wholesale Society undertook a publicity campaign to encour- age the natives to build up stocks of shark fins on each island. The total production of 7,660 lbs. for the whole archipelago in 1950 repre- sented a value of 383 pounds sterling, and shows the small importance of this resource (Table XIX). Considering the irregularity of the Chinese (I) Trochus shell does not appear to exist in the Gilberts. The name baeao, not baio is correct for the usual kind of mother of pearl shell - the name katati is that of a razor sharp shell which is too thick for commercial pearl purposes. Te katai is not known amongst the Gilbertese but may be a local corruption of katati. (Note by the Administration of the Gilbert and Ellice Islands Colony.) (2) Before the last War, a trader (MacArthur) sent samples of sponges to the United Kingdom but failed to find a market for them. After the var, a limited number of samples of sponges from Marakei were sent by the Gov- ernment to Australia and the United Kingdom where firms (including the In- perial Institute) reported that the samples were of similar quality to the West Indies or Red Sea sponges, but would only be suitable for industrial use. Several firms requested 400 lbs. or more of samples, but, despite the efforts of the Colony Wholesale Society, the Co-operative Societies Officer and Tangitang, the people were too disinterested to supply so much as one pound and the scheme was dropped. (Note by the Administration of the Gilbert and Ellice Islands Colony. ) me LO= market and the fact that it is closed for an unpredictable period, the mar- ket for this product is at present limited to the simall Asiatic population of Ocean Island. This outlet, while sufficient to absorb the quantity pro duced now, would no longer be sufficient if shark fishing was intensified. (1) The leaders of the Co-operative Wholesale Society, thinking that the Chinese market might be closed only temporarily, wisely advised the Gilbert: Islanders to keep all the fins of sharks occasionally captured so that a stock could be built up against the day when business should reopen. We saw shark fins in some islands but in a very bad state of preservation. The natives lack the most elementary knowledge for the processing and preservation of this product. Shark fins are sold rather than thrown away when a shark is caught by chance, but the natives will not go shark- fishing deliberately. CONCLUSION Our study of marine resources has shown the great variety of organisms concerned, whether derived from the ocean or lagoon or found on the un- covered flats. It has eimphasized how easily the Gilbert Islander, alone or in small family groups, may find every day a substantial portion of his food, and often quite considerable quantities of fish in return for a negligible effort. We have seen on the other hand that fishing techniques are always very simple and even sketchy and that, with very few regional exceptions, fishing is not practiced on a community basis, which explains why the means used are limited. A suggestion made to fishermen for improving some of their methods, such as adding a purse to their seine so as to jose less fish, or using longer seines, is always answered by smiles and by the argument: "We catch enough fish as it is!" We have mentioned in the paragraph on ikari (Albula) that the natives cannot be expected to try to regularise their consumption by spreading it from one full moon till the next, although this could be done by improving preservation processes. The natives prefer to consume everything at once, since between these dates they are certain to catch or gather other foods, some of which are brought by the sea at fixed times and sometimes in very large quantities. (I) It was the Government which undertook the campaign to encourage the shark fin industry, through District Officers and Co-operative Societies on Islands. The market is not at present limited to Ocean Island. Colony Wholesale Society is supplying Australia in a small way. (Note by the Administration of the Gilbert and Ellice Islands Colony. ) -141 CHAPTER 12 DOMESTIC ANIMALS Pigs In the vicinity of almost every village and generally in the shade, one may observe pig enclosures built of coconut logs or of landing strip matting. In some southern islands, pigs are not always kept in pens but tied with a rope to a pandanus tree in the bush and moved from time to time. Others are left completely free.(1) The pigs kept in enclosures receive an irregular and variable number of coconuts, and some fish waste or unusable food scraps (unusable parts of pandanus fruit for instance). They are also given low plants such as Boerhavia, Thuarea and also the leaves of shrublike plants such as Mes- Serschmidia and Ficus tinctoria.(2) Very rarely, this meagre pittance is supplemented with toddy, which gives a remarkable improvement. The weight of these pigs varies from 40 to 60 lbs. at six months. They sel- dom exceed 150 lbs. after one year. A few two-year old pigs may reach 200 lbs. The average purchase price (on the hoof) is 8 pence per pound. (3) It is a common saying that the number and condition of the pigs follow roughly the alternation of periods of abundance and of scarcity of coconuts. But we observed two-year old pigs almost everywhere which, although rather thin, were not in really bad condition, and the country had just been through a very hard drought which lasted two years. Actually the Gilbertese is quite incapable of seriously raising pigs, although this might provide him with some income. But the sale of copra constitutes a resource which makes the people neglect all other activities, pig-raising in particular. A very severe slump in copra prices would per- haps modify their point of view: this possibility will be examined in the recommendations. The 1947 census gives a total of 6,326 pigs for the whole Gilbert Group, or an average of 0.23 per inhabitant. But percentages are highly (1) Pig owners who allow their pigs to run free are liable to prosecution under Island Regulations. (2) Pigs receive probably on an average two coconuts a day but the main pig foods are: te boi - Portulaca oleracea; te wao - Boerhavia diffusa; te mtea - Portulaca samoensis of which two handfuls a day are given to a pig. (3) The purchase price mentioned is that of Tangitang to island sellers. The meat is sold at 1 shilling, 8 pence per lb. Gow 2 shillings), after freighting, handling, loss and profit charges. (Notes by the Administra- tion of the Gilbert and Ellice Islands Colony. ) -142- variable from one island to the next. Where marketing facilities are found near some village, the number of pigs is higher. Thus in the village of Eita (Tarawa) we counted 42 pigs for 23 families, giving a total of 115 persons and an average of 0.36 pig per inhabitant. Poultry The Gilbertese give hardly more attention to their poultry than to their pigs and they seldom eat chicken. Besides what the fowls can pick in the village, they find in the bush seeds of Fimbristylis, Portulaca, Phyllanthus and generally of all the grasses. Besides at each low tide they go on the uncovered flats both on the ocean and lagoon sides, and find animals (copepods) and seaweeds, although in small quantities. The absence of sargassuim, Cystophyllum, and generally of all the large algae which in other areas are left in abundance at the high water line, is not favourable to the chickens of the Gilberts. The only seaweeds reaching individually a large size are Turbinaria and Halimeda, and their texture is not very well adapted to sheltering crustaceans. In villages providing poultry for an administrative centre, as is the case in Tarawa, the chickens are fed grated coconut and, when fishing has been very productive, some fish. In the village where we counted the pigs the poultry census gave 205 for 115 persons, or an average of 1.78. But like many other averages in these areas, this figure does not mean much since one family may have 40 hens while another has only 2 or 3. The weights recorded are more inter- esting and indicate an average of 1.05 kg calculated from 20 live chickens. Note: In some islands such as Abemama and Aranuka, wild cocks and hens are quite commonly found in the thick bush. They appear to be of the same race as the village birds. -143- CHAFTER 13 QUALITATIVE STUDY OF GILBERTESE DIET General Considerations Some documents have already provided useful details on the diet of the Gilbert Islanders, e. g. Turbott, 1940. These undoubtedly give valuable information on the nature and quantity of the products eaten, although data on quantities are only approximate, as the authors point out. Information on quality (calories, vitamins, etc.) can be based only on studies made outside the Gilbert Islands. No detailed study of the Gil- bertese diet can be undertaken until analyses of the foods actually pro- duced in these islands, or at least of a number of them, have been made. If we rely on those made in America, in Fiji, or elsewhere, for corresponding products ,we may introduce errors, The comparative Food composition tables of the Food and Agriculture Organization are not based upon the products of the low coral islands. As the authors indicate, the composition of each product may be greatly influenced by botanical varieties, climatic con- ditions, cultivation, or preparation, and many other factors. While re- sources from the sea may not show much variation, those from the land must vary greatly. The poor soil, the scarcity of water at certain times and, generally, the special environment of low coral islands must have areat influence on the quality of animal and vegetable resources. Besides, certain common items of consumption are strictly Gilbertese, and analyses of them are indispensable if we are to get data based on the total available resources. What do we know of the real value of baitari, @ jelly-fish eaten every month, of the value of kabubu and other prepara- tions? Analyses of such foods would be the most useful work which could be undertaken by a dietician. Such analyses could obviously not be made in the field, for lack of specialized laboratories, but samples could be sent to research centres with every chance of success provided an expert dealt with their preservation, preparation and conditioning. Air transport would greatly facilitate such research. It wust also be pointed out that estimates of potential production and of average consumption are either fragmentary or inaccurate. It was difficult to avoid this. To obtain an idea of the native diet, one had practically always to study people who had been some considerable time away from their normal surroundings, es for exampie Goverment employees, medical orderlies, domestic servants, or other "exiles". For some of our information we ourselves were unable to avoid this facile solution. In- deed it must be recognized that any enquiry into the total food supply of the natives in their own environment encounters many difficulties. For one thing the Gilbertese seem to experience a certain shame-faced embarrassment in stating exactly what they have eaten each day and, to an even greater extent, in eating ia front of one. Once this embarrassment has been overcome because he has become used to one's presence, it will still be necessary, and this is the second difficulty, to live in very close contact with one or two families for long months (something we could not do). And eveu then one would not have a strictly accurate idea of the Vie complete diet. In addition to meals takeu "at home", the Gilbert Islander,in point of fact,consumes other food the nature and quantity of which cannot be checked. According to the circumstances of the moment, this may con- sist of coconut embryo or meat, eaten while the nuts are being opened, pandanus fruits, berries from small trees, or even marine organisms eaten raw during hours spent on the shore at low tide. And yet - and here we have the third difficulty - it is almost impossible to follow the daily comings and goings of any one individual. It must also be noted that certain foods are eaten on one island and not at all or in very small quantities on another island. Thus on Nikunau, the aatives seem to like sea birds such as te io (Anous sp. « white-capped noddy tern) which are not eaten elsewhere. It is, therefore, difficult to try to fix standards of cousumption for people whose activities and behaviour show such marked individuality. Our modest contribution to the study of their diet will include, in addition to general information, some figures compiled with the greatest possible accuracy. Table X gives the census of a village of twenty-three homes, table XI, gives the number of Cyrtosperma plants owned by twenty-seven families. Table XIII shows the quantity of imported goods which ten families from one village bought in a month, with the addition of estimates of the average consumption of rice and sugar. Family no. 1, with 4 members, who had no toddy, consumed 3 oultces of sugar per day, while all the other families who could drink diluted toddy, used much less sugar. Table XIV shows the different items which constituted the meals for three persons for seventy-five days at a stretch, and the number of times that these items were eaten. The three persons in question lived away from their native village, but not far enough for them to be debarred from obtaining provisions from it. The fact that they were employed and had regular wages, enabled them to buy imported goods frequently and obliged them to maintain a certain regularity in their food habits: they had a light meal, at 7:30 a.m. before going to work, consisting of toddy diluted with water, of sweetened water, or again of tea, sometimes accompanied by bread, less frequently by rice. A second and larser meal was taken about 10 o'clock, and consisted of babai or rice with fish or European canned food. The evening meal was similar ii importance and type of foods. In addition to these tables we have drawn up three lists of food products taken from the "Food Composition Tables" of the Food and Agri- culture Organization (October, 1949). Table XV shows the value in calories, proteins, fats and carbohydrates of sea food corivesponding ap- proximately to what the Gilbert Islanders can procure locally. Table XVI shows the same approximate values for the main items which they can pro- duce on the spot from agriculture and stock breeding, and table XVII for the main imported goods they can buy in the co-operative shops. ay iy Local Products Bearing in mind the constant availability of the various products of the coconut palm, and the diversity of sea food at the disposal of the Gilbert Islanders, we may conclude that these two sources of food supply would in themselves, be sufficient to prevent serious deficiencies, par- ticularly of vitamins. Many other items, fresh or canned, are added and still further increase the margin of security. Several of these are said to have but a very mediocre food value, but their variety and quantity automatically compensate for this. For instance pandanus keys give 26 calories per 100 g, but foods made from pandanus, such as kabubu, are use- ful more because they contain a large proportion of grated coconut which has a value of 606 calories. In other cases the low value of certain products is compensated by the very great frequency with which they are eaten. Other valuable foods such as coconut meat from the gelatinous stage to the copra stage are also eaten in large quantities, even betiveen meals. Finally, there are some very important products which play a pre- dominant role in the dietetic balance, for example, toddy; if it is eliminated from the diet obvious end undeniable signs of disequilibrium follow, after a length of time varying with the individual. According to Dr. Bray the food value of toddy is: Proteins 0.32, fats 13.0; caloric value per 100 g., 54. It is still an important part of the diet, but it is characteristic that every person or family whose finances permit shows a marked tendency to abandon karewe and replace it with sugar. In many respects this is a regrettable substitution. Many authors have drawn attention to the frequently drastic deteriora- tion in teeth brought about by the consumption of sugar among people whose percentage of decayed teeth had remained insignificant as long as their diet had remained Natural. The same authors usually agree in recognizing the role of white flour in the progressive weakeMNing of teeth. The great majority of Gilbert Islanders have superb and healthy teeth, so it must be concluded that this majority has not yet slipped into an unbalanced diet, thanks to isolation and the small quantity of imported products that are procurable. We may add that the powerful mastication which is required by people's diet is an additional factor which tends to preserve their ex- cellent teeth. It takes much chewing to cope with certain dried fish, coconut meat, babai tubers, or tuae, or to shred the edible mesocarps of bunai nuts or pandanus keys. We have dwelt upon this question to draw attention to possible re- percussions of the substitution of sugar for toddy, were it to become general. This is not the only danger, however, and we aoted that the preparation of traditional dishes derived from babai, breadfruit or pan- danus, and the collecting of sea food are totally neglected as soon as any money becomes available in a household. -146- SL 9:0 Ot = 08 1 e620 0-1 086 “sqT - zedng C. We SOP 6-8" st T “9°0 <0°o Cee) Se ord "sqT - 9oTY OST Jad reSns pue sotr Jo uoTydumsuooa ATtep eBereay = Ot =-9 © Ot = me (i CRE = Ory "sqT - zedns Zl clea i =O 19 Se GE hs as es 019 “SQT - d0TY F UBD = Z = = ~ - - q Z 4 'ZO ZT - MTTW peszepaog = ~ zr - - - - ~ - re steed ued °ZO OT - - - - - Fe} - 2 - T Me1S aTqeyeToa pue 1eAI - 1 2 2 = - - 7 Z € Sones o1BUI0. UT SSut17z0H - - - = - ~ - - - PA ued *Z0 Q - UuOUTeS os = ~- a @ 2 cd 2 tA ues *Z0 OT eon Ay, Tpowmuog = = See T - - - 5 Cera Sa - . squejut G a G € G € 9 On 6 + suosied JO ‘Off Ob oe ee ee ; AT weg “TTO1V enerey, vO a8eTTIA e UL soTTruer uaa Jo yora Aq YAuCtTt avo UT YYSNoOqG saTzTpovmMod peyzodmt jo jUNoUy IIIX Wiadvb TABLE XIV Number of times the food items listed below have been consumed by 3 Persons (A, B and C) during 75 days A B c Mature coconut meat (te ben) 13 32 te) Unripe coconut meat (te moimoto) 3 rs at Toddy 20 116 83 Toddy molasses (te kamaimai) 30 2 1 Cyrtosperma (te babai) 21 19 43 Food based on babai Te tangana 8 i. Te buatoro G Te Bekei 9 4 4 Breadfruit (te mai) it 1 Dried breadfruit (te kabuibui ni mai) 2 bl 1 Dried pandanus fruit (te tuae) . i 5 Pawpaw, papaya ab if Squash 3 1 Poultry is } Eggs 2 Pork 2 aL: Fish, fresh or dried 78 88 qe Crabs ay Shellfish 2 al Octopus 1 Rice, polished Th 112 100 Fish, canned ay) "i on (Herring in tomato sauce) Meat, canned ot 31 32 Bread he 32 50 Food based on white flour Te Katiobuki 10 ge 2 Te Tonati 2 2 Pancakes h in ¢ Tea 57 ETE 32 Sweetened water 83 54 106 Milk 5 2 Jam 1 B Coffee a 1 Potatoes 2} Arp &£iaA ) Te 09 Oh eToyN usyeo YSTI - pet4zq LEeS™ 9) qT GS TOE vayes seuog “USTs TTeUS 9€2 6) qT 6S T9E papreostp sauog “YysTy VTE G€g spoyeety ATRu0IIS 0) OT On 196 uazes souog “UST TTeUs HE? 0 == Ole Ot 19g ‘PePreosEP souog “YysTy eBreT C2 “$191 C0TE :pezeety ATO7e1epop *ssUTIIEY peLTes 0 TT Te 69T ‘uayee sauoq “YsTJ TTeUS ote pue poyous fsuT.1q UT Oo TT T2 6QT ‘“pepszeostp sau0g UST] SS.1eT lige ‘Tereyoeu ‘uouwpes 2peyeeay ATIUSTTS ‘ssutpies ‘sSuT710q 1ey UT YoTr setoaedg petup 9°? Hee 9°hT ge UOTIOUTISTP Ynoyyta ‘Y70g Of2 "Oyo ‘sueTO ‘stassnm ‘s.1948h0 67g G°T €T 0g SYSNT TON 622 “ote ‘qez fustyAcro §si3a1sqoT Ord Tae eat COT suesoeysnig 922 ysery ‘SYSNTTOM pue suveceysnig 6) E2S Q°eT cet aTOyA - 0 BS Q°QT FAS PSI8TITS - UOTYOUTI -STp ynoyu2TA setoeds [TV 922 ‘USTJ JOJeM USatTT “wee 4 ees fyorsd vos “‘qeT TN ‘sepnoe11eq ‘syzreys fe) (Gor) 6T OT TOU - fYSTI VET] :sopnTour ©) G*2 6T OT PESTITI - setosds 19410 427 1 2U4TES,, @) S°0 4° OT GL STOYA = ecc w7sno,, “yooppeq ‘eyey ©) Sie) OT GL PSteTITS - SotToeds Ie[IUTS pue pon ced ysery ‘usta 4 4 % 3 QOOT ted TSqumyt S7.TEMMIO| soyerpkyoqre) )§=6 seq 2S SUTAQOUg SOTIOTSD uoTIV.dT19seq pue wWeIT Tet1eg ; “SIITUBSW TAYZO OU SsAeY pues (646T Ova) y20d9.2 OVA 24u4 JO esol. are szaqumU TetT.1as aul ‘Wet YoeSa 101% sarerpfkyoqres pue syey ‘sutTaq01d jo esequsoiad ayy ‘sueiS OOT ted satzoTes jo azaqumu ‘pooy vas pues YST]I Ute .1a0 JO uoOTITSOdwoo |Yyy Sutmoygs AX @Tdvi EEE ) ¢ re) OTE uvazea souog “YSTI TTeUS Cug ) G re) OTE peprzeostp sauoq “YsTy a9zeT] one :peqeety #£TSu0I1S @) € 9H €z~e usye2 souog SYST TTeUs Ti? ) C On Eee pepreostp sevoq “YysTy avreT OZ >peyeer2 fTay,eTSpoy ) ra G° Lz GET uayea sauog ‘YSTI TTeUS 6€2 0 ro G' Le GET pepreostp seuog ‘YsTy ere] gta a13tem ‘meeIg pervert. ATIUSTTS eos ‘poo ‘yooppeH 2487 ut rood satoads % of, go, S001 12d TAquMayt S71 UemMIOD sezerpsyogrzeg §=6s2eq SUTSLOIg SOTIOTED uOTYAT.1OSaq pue wegt Tetszag (uoTPenuTyUC)) AX Wav yserty ; 6'0 Ueealan Too €OT - 0982S Teanzeu - svvo us Ct2 “smers O0S‘T (setzoseze. [Te) pareeto 76d aTqtpe you ‘payontd - yySTOM 0 ieee a Oe 002 Teo} ‘useyoTYO pexyontd 402 SOSSBDICO WOT] PoWUIT.12. 1eq ) 0°69 iG: 91g set1oveze. [Te “nay “st tO¢, 692 ‘Sy 1G sseored ae ‘Sy 6) SATT “IUSTOM 0 O°SE ie OLE sseored - ueetT ‘¥x10d LOT TOG jE (6) 9°0 Sa Be (ekeded eotszep) iwediveg QuT ion 7°O Oak G9 (eoTreo SsnoTA) STA THT (‘dds sndzes07.1V ) yTNIzy ASxzUVOUI,, Ge le 4°O TE YQ ‘qrnig yore “ytnzgpeoig . gtT S°6 G'0 g°0 TH (etuOUTT °9) suoweT ten (dds esnw) petstoeds G°92 i7°O 2°T Cort -un ‘sutej,uetTd pue seueueg 6IT *UdaIT us1e9 S4ITqmmMoond ey.O pue (‘dds aeodeyTqinong ) ‘smnorzem ‘saysenhs reumvms Gag TG ZOO Ge ‘users ‘saysenbs ‘sp.tnoy Gui: S.Inqgeut - 9n4 eu S4Tqinono UsYUM usyZeS S4ATqInong Ret €°o Sar ee Jey10 pire sutydumg qTT -o10 Seootdey fyoTe1s UTOD £4OO.TMOIIY 6°98 EO: (S210) e9t kip famd ‘sayoreys Sa ene ran) 6°T OT ("dds etseo0To)) o71eL On T’9 1S) 919) eae TE (‘dds eewod~) szooys eoulodyt 96 (sez27eq, €'l2~ ae) €°T AE eawody) seo.e.Og 7908S of qnuoso0o ‘guresas *y071700 ‘dATTO O OOT fe) uists) STTO sing Ns? OT GT a OST yor sinyzermt JO Leo} G9 get HE ot Ece yeow ecty 79 $1.nU0s00 of 9 9b SOOT ted T3aqumyl sa. USwUIOZ soverpAyoqre) s71eg SUT9VOId SdeT.10TeD uUOTLATIOSag pue WaT Teg : “Sutuesem 1ey4O ou aaey pue 4yx10dezI OVA 34 JO asolyy are SLTaqumMU Tet1as sy ‘Wat yore roy saqaerpAyoqreo pve syez ‘suTezorud go avequaozed ayy ‘sureqS OOT ted satzoTeo yo zaqumu ay. ‘pooy vas pue YST]I UTeLIAD JO UOTZTSOdWOD ey. Sutmoys TAX WIV "6H6T 190GI1N], UT ‘aseastq 07 AtysvuoTzeTaYy pues sepToUeToTy -ad oT}JO}ETC Ss,ferIG “M “DN ‘AG WOTZ poemMorI0g UsEeq aAcY Vea pauueo AOS sarNnBTq (T) 0 G22 G°S2 qTe zea pouueg (T) *psaouer (%og jnoqe) STTSsys JaLFe suesq voo00| (0°02) 0°0S (S) NaeS *pous7.990/S -un ‘fkreuTpazo farepooo"g 92 ‘ore faazefnap ‘azeppeyd ord O°TE G2 LEE “YT TW aToyn ‘passac0id - asaoyQ 902 wopsuty peyYtuN IOS prepuerzs (WAS omens Z°8 Qce ‘poueza0ns ‘pasuapuod fstoya ‘peuueo ~ (S,400) NTTW €9¢ “mOpsuUTY pe1TUQ pus epeueD : *s9289¢ pe1TiQ 1Or parepueys 6°6 Son Ore OCT ‘paveyaensun ‘pasuapuod ‘atoyn *petues - (S,N09) TTI o92 O12 “Sty aT1ano ‘stessnu ‘szoeasko O'2 OVE oye de Ly peuueo “sysnTToM 642 -oya‘qezo ‘zaysqoT ‘ystserp @cr SFE Cat 26 pouueo ‘suesoei1snig O42 "483 UT 100d seTtoeds 6) 0d O° Td 90T - THO JNOYPTA perueco ‘usta Le ‘Vel UT Yotsr satoods (OE (re 0°0¢ COT - TEO PNOYZTA pouteo eur One soTstove1e. O'T O't2 0° dd HTE Tre ‘TtO UL pevUeO YSTA Sie 6°21 T£2(0) 9°0 ot pouazaonstm faotnf avuet9 GOT *sqTNIy peuued jo (zesns Suppntoxe ) VYBTOA You 04 sat Tddy 0'6 Zz'0 E016) 9€ Teaqsues ur - pouues *sqtntq 4O9T 0°6 a0 4° Oh (edeo unt{Ty) edra ‘suotud Ld 6°eL 1 O) L°9 O9£ SoTY S4TUM aT SOUS Fae a 6°0OT OLE (qesym) ‘aaTyA SamoTa € 9) 9, % wd OOT ted TAqumnyl szucmMoD saxverpAyogsre) s1eq suts.o071g SOTIOTED uoTIATINSaq pue mWs1T Tet4tesg ‘JuTUuvalT TauZO OU aABY pues 41OdderI OVA ayy JO |asoy. ate STequnu TeTtes su, *We4T yora z0y saqyerpkyoqrtes pue syez ‘sutaqyord go aVequaored oyq 6 9 OOT ted satT1toTeo Jo Jequmu ey4 ‘sdoys ayy ut «nq ues aaTVeU ay. Yotym syonpord peyrodwmt jo uotyTsoduoos ayy SUTMOUS ITAX WTavL TABLE XVIII Estimate of vitamin contents. Vitamin content calculated for edible por- tions of 100 g of selected food products. Figures taken from the tables prepared by Buchanan, 1947. Thia- Ribo- Nico- Ascorbic PRODUCTS ' mide flavide tinic Acid Bel B-2 Acid C P-P mg me mg mg Cereals Biscuits - dry Oe dk O Biscuits - sweet 0.05 O Bread - white Gia: O Flour - white - sifted to 70% 0.06 0205 0 O Rice - husked at home or slightly j bleached. 0.2k Ord) AO @) | Roots and Farinaceous Roots . Arrowroot 0 (0) (0) 0 Bananas 0.05 0.06 O50" 10 Breadfruit pulp - cooked | \| Breadfruit pulp - raw (0.2) - - (15) Plantain, green 0.05 0.06 On5 10 Animal products, Proteins - Fish Crabs, crayfish, spiny lobsters - 0 O e) Fresh, with oil 0.05 OLR BLS: 320 * Fresh, without oil (medium) 0.06 0.35 3.0 *. Freshwater and sea-water shrimps - 0) 0 @) Salted, dry and hard 0.06 OL 3h Tae) % Sardines in oil 0.02 0.2 4,0 CO} Shellfish 0.05 Om 4.0 * Tortoise - - - - Meat and Esgs Begs 0.14 0. 34 * * Fresh Pork, lean 1.04 0.23 5.0 * Fowls, plucked and cleaned ORaKS OVE 6.0 * Fats Butter O 0 0 @) Vegetable Oils: ex. coconut oil and peanut oil 0 @) O @) Nuts Coco - ripe flesh, fresh. Oya 2 (0.4) i - milk, ripe nut. * * * * - water, ripe nut 0 ~ ~ 2 Green coco - pulp Found only in green nuts. TABLE XVIII (Cont'd) Thia- Ribo- Nico- Ascorbic PRODUCTS mide Flavide tinic Acid B-l . B-2 Acid C P-P mg ng mg ng Fresh Vegetables Beans, these are all eaten green. C1064 sOnde O25 20 Cucumbers 0.04 0.05 0.18 10 Leaves, fresh, dark green, such as; leaves of Cassaba, gourd, sweet potato, Chinese cabbage, young pawpev, taro. Out Sun. One5 0.85 100 Onions and shallots 0203) 1 Ont OLY LES Gourds or squashes 0.0) (0:05 Ong 5 Unpeeled tomatoes OrO5 Oro Or? 25 Drinks Cocoa powder Oe 10.27 ik ~ Coffee grounds ~ (OJAIE Rae ~ Tea leaves = 130 6d - Note: * contains such a small quantity that it may be ignored in estimating the food ration. 0) nay ( ) probable value. Earlier, Grimble wrote: "When times are good, they live ‘off the land'; in times of famine they live on reserve supplies". We may para- phrase this by saying that nowadays, in times of wealth the Gilbertese rush for canNed goods, rice and flour, and only poverty forces them back to the balanced diet of local products. It would indeed be uawise to generalize, but it is characteristic that the consumption of many imported goods tends to become an unbalancing factor, instead of an addition or supplement to the diet. The Gilbertese still consume toddy molasses (kamaimai), but, actually, they never eat as much of it as they do of sugar, which they take by the spoonful, with real avidity. (1) The balanced diet obtained from some rich foods and the diversity of others may be the reason why the natives fail to use other resources easily availabie to them. In other words, if these people were really suffering from an ill-balanced diet, they would not disdain, as they do, fowls, eggs, and the fresh leaves of certain plants, which are freely eaten in other regions of the Pacific. The very small quantity of pork - the only source of fat meat - in the annual consumption of the Gilbert Islander is obviously very ) largely compensated by the protein contribution from fish and many other marine organisms, such as crustaceans and mollusks which are eaten in great quantities. In this connection it would be very useful to have the results of analyses of the edible jelly-fish baitari. The kolukale of the Tonga Islands, another edible jelly-fish, belongs perhaps to the same group of Carybdeidae. To return to that most important food supply - fish, it must be noted that the Gilbert Islander uses almost every part of it and that he is particularly fond of the liver, very rich in vitamin D. Often he will not even wait to bring the catch back to the village but will cook on the spot, liver, hard and soft roe (milt). Sharks' livers also are always used except, of course, when they show certain signs recognized by the fishermen as indicative of poison. We have already said that chickens are very rarely eaten. (2) It would be wrong to infer from this that the Gilbert Islander does not like (1) Although the Gilbertese have a liking for sugar, it is not as marked as their liking for fat, which amounts at times to a craving. (2) One important reason (not mentioned by Dr. Catala) for the Gilbertese not making a habit of eating chickens or eggs is their desire for a large quantity of food at each eating session. A whole chicken or at least six egys would be required to satisfy them. As the local supply of these com- iodities is insufficient to satisfy such gargantuan repasts, the Gilbertese do not bother much with them, although in the last 4 years more interest in this form of diet has been evident. (Note by the Administration of the Gilbert and Ellice Islands Colony). sire chicken. Sufficient proof of the contrary is the rapidity with which he disposes of every scrap left over from a European meal. But on the ad- mission of several, this neglect of fowls is due solely to the trouble in preparing them: plucking, cleaning, and lengthy cooking which must be supervised. They will take all this trouble only for the nourishment of old people, very young children, and invalids, who are given chicken broth, or for certain special occasions, or when a visitor has to be entertained. Curiously enough, eggs are not prized either and are eaten almost exclu- sively by children. They are not used in any of the dishes in which the Gilbertese might profitably include them, such as katiobuki, tonati, etc. There is the same neglect of pawpaw fruits which are also reserved for small children. Yet the papayas of these regions have quite an outstand- ing flavour. as for bananas, they are a luxury and form only a minute proportion of the diet of the mass of the community. Pork is very rarely eaten, not because it is disliked, but because there is little of it and it seems to the Gilbertese quite useless to take the trouble to raise pigs. As one native remarked: "We have already more than enough to do feeding our babait" It is deplorable that such a small quantity of pork should be included in the diet of the Gilbertese, in view of the wealth of high-grade proteins it contains, particularly in the liver, kidneys, etc. which are also rich in vitamins A and D. Imported Products Rice is eaten in increasing quantities throughout the islands, and it is regrettable that this is solely polished rice. As a result of our conversations with the Gilbert Islands authorities, we discovered that the native is now so accustoued to this form of the product that it would be difficult to make him accept it in any other. We are not so pessimistic and, rightly or wrongly, we think that if the Gilbert Islander had nothing but a less highly processed rice, for lack of better he would begin to use it, and would gradually become accustomed to it. It would be necessary to find a rice with a lower processing rate than bleached and polished rice, perhaps even cargo rice. The ideal would be to deliver for consumption either fortified rice, or simply rice still in the paddy state although we cannot visualize the Gilbertese family mak- ing the effort to husk imported paddy.(1) (1) Consignments of semi-polished rice imported by the Colony Wholesale Society have not been popular. Dr. Catala's other comments on imported products are correct, but the quantities imported are small, i.e. some 20 pounds rice, 20 pounds flour, 12 pounds sugar per head of population per year and a high proportion of this is consumed by Headquarters staff. It is amongst the highly concentrated population on Betio and Bairiki that the danger of an unbalanced diet is most to be feared. As the ordinary Gilbertese only does the minimum of cultivation for himself, it is diffi- cult to encourage him to cultivate for trade. Brown Malayan rice has been imported into the Gilberts, but was only purchased by the people when no other rice was available, and even then with much grumbling. It would take time and the complete stoppage of white rice supplies to accustom the people to such diet. The islanders show no interest in food values, and compulsion will be necessary if more nutritious foods are to be introduced. (Note by the Administration of the Gilbert and Ellice Islands Colony.) eae = We can simply express this wish: that an excess of polished rice in combination with an increased consumption of too highly purified export flour may not one day coincide with a period of restrictions affecting products rich in vitamin B, like toddy. We have already seen that there is too great a tendency to substitute sugar for it. The results would be an unbalanced diet which might lead to more or less serious cases of beri-beri. Toddy making is neglected.as soon as there is enough money to buy sugar. From the most reliable sources it was learnt that every load of sugar arriving on an island is bought up almost at once, until every one's available funds, earned from the sale of copra or possibly some handicrafts, are exhausted. Canned Goods: We shall not enumerate here the varieties of canned goods which may tempt the Gilbert Islander. Fortunately canned meat is most often eaten. Table XVII shows its high value in calories, protein and fat. As a result of our calculations we find that the “average Gilbert Islander" eats approximately ten cans of meat per year. Wage earners eat much more, and the figure of two tins per week is often exceeded (see table See Next to meat the most highly valued item is herrings in sauce. In stores in the southern islands, we even saw cans of salmon, but they had not proved popular and had been in stock for two years. We were not able to find out whether this was because of their price or because the cans were. swollen. (1) It mst also be pointed out that the natives are very fond of fruit in syrup but do not often buy it because it is too slight a return for the money expended. "Navy biscuits" are much liked. One more word on the subject of flour. It is very often used in many culinary preparations which could do without it, because it makes them easier to knead. Very surprising mixtures indeed are encountered, such as sweetened milk with meat. We must, however, make it clear that it was only in very exceptional cases that we saw cans of milk in the shops of islands other than Tarawa. The Gilbertese like tea and coffee; old people drink tea morning and evening with sugar, not so much because they like it as because they have no one in the family to collect toddy. Coffee is rarely con- sumed, prices being extremely high. Note: Generally speaking, it is estimated that the minimum ratio of carbohydrates over fats should be greater than 1/!-, and that the minimum weight of carbohydrates should be from 50 to SO g per day. Such a figure for the Gilbert Islanders has yet to be determined, if it were known, it might provide an explanation for their marked teudeicy to eat more and more sugar and flour, always supposing that this tendency is not only the result of greediness. I: it turned out that additional carbohydrates were really necessary, it would be essential to add a complementary amount of vitamin B which, in practice, would not be easy. (1) The only reason canned salmon is not bought is because of its price; the people like it! (Note by the Administration of the Gilbert and El- lice Islands Colony. ) . -149- Conclusions 1. Contrary to what is generally thought, the majority of the Gilbertese people do not suffer from malnutrition or, stated in another way, from an unbalanced diet. Useful conclusions should be drawn from the fact that it is precisely those natives who have remained most natural in their food habits, and generally speaking, in their mode of Life, who are least affected by the difficulties resulting from drought. The Gilbert Islanders - particularly those from the southern islands - experience, at least once every six or seven years, a period of intense drought which may last two years, but the ill effects of which really begin to appear only after the first ten months. These people are thus compelled to adopt a frugal diet which often borders upon the restricted, but does not appear to affect them very greatly, for the majority of products at their disposal in normal times, except coconuts, are still available. They are protected against serious famine by the continual diversity of their food supply anc more particularly by the continued production of toddy. It is noteworthy that the people of the southern territories, more severely affected than the others, are the ones which take the greatest care to build up various food reserves from pandanus (kabubu) and breadfruit (kabuibui). It is on these islands also that the greatest care is taken to preserve desirable varieties and to replace trees as they are depleted through age or after a drought. 2. In speaking of the Gilbertese there is too great a tendency to con- fuse under-nourishment with malnutrition. They are subjected to the one without, however, suffering from the other. From this confusion there has arisen a kind of famine myth which does not correspond to fact. The Gilbert Islanders themselves, while recognizing that they occasionally experience periods of serious quantitative restrictions, never e:agzerate the gravity of these, and tend to smile if one commiserates with them. Moreover, when we asked them, after two years of drought, how it came about that they were still in such gocd physical condition, they gave us this unexpected reply: "Here, food is so easy". It would, however, be inadvisable to ignore the lowering in activity and resistance to which their constituvion is subjected as a result of such restrictions when prolonged. The so-called "depletion" of fish around certain southern islands would really seem to be attributable to a general state of asthenia among the fishermen. The Gilbert Islands' Medical Of- ficers mentioned the almost complete absence of beri-beri but emphasized the fragile balance of these peoples. It seemed to us, but only on the island of Arorae, that the physical state of a rather large number of people reflected the particularly poor condition of the vegetation, especially the coconut palm. Everywhere else, although an era of privation had just ended, the general appearance of the natives was far from dis- tressing, even in the southern territories. There, as in the rest of the archipelago, many available sources of supply were neglected, such as chickens, eggs, pigs, proving that no need for them was felt and that the diet of these people, however monotonous it may appear to us, was still sufficient. -150- We may conclude thus. This race has a imich greater need of balanced rations than of a super-abundant diet. As long as the Gilbert Islanders’ food supply remains what it has been in the past, there is every reason to believe that he will possibly be subjected merely to temporary food re- strictions, not to an unbalanced diet. These periods of restriction will never have on the constitution of the Gilbert Islanders the consequences which would result from a diet unbalanced by the over use of imported pro- ducts (without compensation). Ina way, the poorer these people are, the better the balance in their diet; this paradox inay vindicate the systen of holding back stocks as emergency supplies, as practised by the native co-operatives. In this connection, there is one important point: through- out the entire Gilbert Islands territory, there is not one of these firis for which, generally speaking, "business has its reasons which reason does not recognize". All imported products are in the hands of a single or- ganization, the "Co-operative Wholesale Society". The administrative authorities of the archipelago who have more or less - but, in our opinion, to an insufficient degree - the right of super- vision of these co-operatives, = have here an exceptional opportunity. They can prevent the "stores" from being induced to sell too much of cer- tain products, or products which are imore harmful than useful. They may likewise suggest that preference be given to one commodity rather than another. They will be able to do this even more successfully when deci- Sions can be based on research work conducted in the field, over a con- siderable period of tine, by a dietitian whose presence in the immediate future is imperative. The dietitian should always have precedence over the grocer. \1 (1) Although a Co-operative Society is at liberty to import its require- wents direct from overseas it rarely does so in practice, preferring to buy from the Colony Wholesale Scoiety on which it is often financially dependent. The Govermuent has, therefore, (through the Colony Wholesale Society) the power to limit the amount of imported foodstuffs. For the time being international quotas provide a sufiicient restriction without having to resort to the somewhat objectionable form of control advocated by Dr. Catala. There is a genuine public demand for imported foodstuffs which Government wovld find difficult to resist and which would inevitably be satisfied by some other, and probably less desirable, means should Government insist. Furthermore the Colony Wholesale Society is not a charitable organization and must iake profits in order to build up ade- quate liquid assets to enable it to trade. It is of considerable economic value to the colony since there are no share holders nor overseas invested capital, and every penny it makes remains in the Colony. The present policy of the Colony Wholesale Society is not to create artificial tastes for imported luxuries but to concentrate on the import of basic commodities which are of real value in supplementing local production whilst making no attempt actively to discourage individual Co-operative Societies" orders. It is misleading to say that the administrative authorities of the archi- pelago have the right of supervision of the Co-operatives. Although the Co-operative Societies! Officer is in practice an administrative officer his primary concern is the business success of the Co-operatives and every effort is made to prevent the Co-operative movement becoming a quasi- Government Organization. (Note by the Administration of the Gilbert and Ellice Islands Colony. ) -151- CHAPTER 14 GILBERTESE HANDICRAFTS In addition to various objects needed in their daily lives, the Gilbertese manufacture certain other articles to sell, such as single or double mats not dyed but made of differently coloured straws, children's mats, coconut fibre mats, woven baskets (round or rectangular), various kinds of small table mats, wide-brimmed, finely-woven hats, and swords of hardwood decorated with sharks' teeth (models of old weapons ). Unfortunately, the trade value of these articles is much less than it could be. For example, a place mat is sold for 1 shilling S pence and a glass mat or coaster 3 pence. Providing the fibres were already pre- pared, one could be made in a day. A single sleeping mat costs le shillings. It would take a week to ake one, at the rate of ten steady hours of work a day. We do not think that any Gilbert Island woman would make this effort, even in double the time. For information, we have included in the table XIX the quantities of handicrafts exported in 1950 with their value for each island (sharks! fins are included). These articles are manufactured particularly in the southern islands where, living conditions being more severe than in the north, the Gilbert Islander has greater need of a subsidiary source of income during years of drought. However, as soon as copra sales soar again with the return of favourable conditions, he immediately abandons this additional source of income. Indeed, the greater the production of copra the smaller will be the output of handicrafts; but inversely, handicraft manufacture will not increase at a rate proportionate to the decrease in income from copra. The diversity of articles manufactured varies with the islands. Thus Arorae mainly produces wide-brimmed hats, table mats and sleeping mats; Onotoa, coir strings and ropes, and baskets; Nikunau, various types of baskets and swords decorated with sharks* teeth; Tamana, coir string, belts and mats. The Island of Beru produces everything, while Butaritari, in the north, has concentrated on the production of baskets. Since these various articles should be considered here only in relation to their export value, which is conditioned by their appeal for outside buyers, it may be useful to examine their chances and conditions of suc- cess. Among articles of limited appeal are replicas of early weapons. However pleasant may be the tradition which impels the Gilbert Islander to offer the visitor these models bristling with two rows of sharks' teeth, it is regrettable that he takes such pains to make them. Apart from museums and a few collectors, who specialize in this kind of trophy, few curio-hunting tourists would be interested in such cumbersome objects, which are difficult to transport and can inflict injuries too easily. Articles having a ready sale include table and sleeping mats, coir door mats, and other products of coconut or pandanus leaf fibre. These are both original and useful, with very attractive patterns resulting from -152- TABLE XIX Quantities of Handicraft Products exported in 1950. Mats Fans Swords Baskets Sharks' fins Export value in pounds sterling Island Units Units Units Units Weight (lbs) Pounds Shil- Pence lings Makin - - ~ - 80 5 Butaritari = - - 200 ~ 38 10 Marakei - - - = 108 8 15 Abaiang 9 - < - 165.5 15 Tarawa 1 - ~ - Mas (22 179 7 h Maiana 5 - - - - h, 19 Abemana - - - - - - - - Aranuka, - = ~ - - ew - - Kuria - - ~- ~ 85 5 6 3 Nonouti 6 - - 3 - 2 Tu ING) Tabiteuea il - - “ 252607.5 153 13 6 Beru 536 i z 53 228 62 Me Onotoa - - - 39 JUSTE 23 Bip aiid Nikunau - 7 Ba - 398.5 43 SW atal Tamana 150 = = = 310 68 ney 6 Arorae 221 - = 1S: 1,128 164. els Totals Oks 7 55 310 7,659.5 7192 Aaa Male one ie =e rt AA " Ani A the skillful use of light and dark straws. Fine straw hats are very handsome articles and there should be quite a good outlet for them once the search for markets likely to absorb them regularly has been properly organized. Quality and Defects Many of these articles disploy very fine handwork and vemarkable regularity in weaving. Unfortunavely, little attention is paid to uni- formity in pattern or size. For example, .we always found it impossible to make up a matching set of a dozen or even half-a-dozen place mats and coasters, however great the number of articles of each type brought to us. Conclusions Déspite their indisputable importance, these products can bring only a limited monetary return to be added, for the benefit of the native, to the reserve funds already formed by the co-operatives. They vould not suffice to save the population of the Gilbert Islands, were they one day to suffer a more intense and prolonged drought than they have done so far, and need to buy great quantities of food. (1) The shortcomings of non-matched sets of table mats, etc. has long been appreciated and since the inauguration of the Co-operatives every endeavour has been (and is being) made to remedy this. There is, however, too much splitting up of the work between individuals which inevitably re- sults in badly matched sets. (Note by the Administration of the Gilbert and Ellice Islands Colony). -153- (1) CHAPTER 15 CO-OPERATIVE SOCIETIES It is not within our province to go into details of the strucwure and accounting of the co-operative societies as they are described fully in the annual reports of the administration, from which much of the in- formation below was obtained. The principle of these organizations and their various objectives are the only matters of interest here. It is, however, useful to review briefly the successive stages of their develop- ment. In 1931 Mr. H. E. Maude was responsible for supervising the formation on the island of Beru of two co-operative societies, known in the ver- nacular as boboti. They were a replica of a society established at Vaitupu in the Ellice Islands in 1926 by Mr. Kennedy and their rapid development showed that they met a real need. However, they were only small organizations; the number of members of each did not exceed two hundred. Their income derived partly from the difference of 10% between the price at which they bought goods and that at which they resold them to members, and partly from the difference of 10 shillings per ton of copra between the price paid the producer and the selling price to the private trading companies. As early as 1934, the Gilbert Archipelago alone had thirty-four of these small societies through which were channelled a very high proportion of the transactions between producers and consumers on the one hand, and the trading companies on the other. In 1941 came the war and the Japanese occupation. The societies were automatically liquidated by the cessation of their activities, and pri- vate companies disappeared for the same reason. The war thus created a new situation. From 1944, the Government, which was concerned with re- storing the economy of the Gilbert Islanders, supplied them with stocks sent from Fiji, and shortly afterwards formed a "Government Trade Scheme" on the strength of a loan of 23,250 Australian pounds granted by the metropolitan Government. The aims of this new organization were to import various consumer goods, to export copra, and to develop other exportable products. Since 1946, however, new encouragement has been given ta the formation, on a wider basis, of a co-operative organization extending to all the islands. As a result of the activities and perseverance of men such as W. G. Alexander and I. G. Turbott, confidence in the co-operative societies steadily increased. These organizers have taken into consideration the different traditions of each group of islands. A committee with its own administration was formed on each island. The "boboti", the former private native companies, and the “tangitang", 2 sort of syndicate, were incorporated into the present co-operative societies. -154- In 1947 the number of members was 5,050; in 1948, 7,898; in 1949, 9,671, and in 1950 approximately 12,000. In the same period the number of customers increased from 6,000 in 1947 to 24,800 in 1948, to 27,500 in 1949, and to approximately 32,000 in 1950. Everyone on an island can be considered a customer. Although only the head of a family is normally a member of the local Co-operative Society, all his family consider them- selves equally members. The continual improvement in the standing of the co-operatives may be attributed both to the system of organization and to a very high copra price level as a result of which the Colony Wholesale Society made large annual profits. The Stabilization Fund is at present (1951) of the order of 120,000 pounds, of which some 40,000 pounds are on loan to the Colony Wholesale Society and 30,000 pounds invested in Australian Commonwealth Treasury Bonds. The interest is 3% and is reinvested every six months. As copra is the major resource, it may be advisable to indicate the percentage distribution of the F.0.B. price paid by the United Kingdom Ministry of Food. The following are the figures given us by the Adminis- tration of the Gilbert Islands. I. Export Taxes 25 II. Cost of handling and loss in weight at the "Colony Wholesale and Island Co- operative Societies" AG III. Commission and profits of the above- : mentioned Societies 12D IV. Contribution to the development fund of the Co-operative Societies and Copra Stabilization Fund Ley V. Net producer's price 30 Total 100% ( al ) (1) The 1952 copra price figures were as follows: I Export Taxes 20) Ol hue 25% II Cost of handling and loss of weight; av Che ent) 3 et. island approx. 3pds. LEAS) Ve 195% III Commission and Profits: Cease O.1Oe 16 Island Society 3.16. 8 UOMO mie 133% Ly YC -SyDevand)C.Wiy Fund Selon. 10:3% V Net Producer's Price 2B ie diay df 314% SH SteN) 100 % (Note by the Administration of the Gilbert and Ellice Islands Colony). * Prices are in pounds, shillings, and pence. -155- The relatively high proportion which is placed in reserve will be noted. The latter could serve as a security fund in the event of a disaster or if the price of copra fell so low that even its transport would no longer be justified. Apart from the reserve funds in the banks, the co-operative or- ganizations already own considerable equipment; two motor boats, six sailing ships, three launches, three punts and seven trucks, to which must be added docks, copra warehouses and miscellaneous equipment. Finally, in addition to voluntary contributions to the reserve funds or the equipment purchasing funds, the co-operatives are responsible for the costs of administrative and executive personnel. In addition to activities directly Se to copra, there is the Colony Handicrafts Co-operative Society 1) which since 1950 has been dealing with various products and manufactured articles - see earlier chapter. Conclusion Apart from the present importance of the co-operatives in the activi- ties of the archipelago, the future role which the accumulated reserve funds may play one day ensures for the mass of copra producers - in fact, for all the Gilbert Islanders - security which might extend over several years. The co-operatives have already anticipated that in the event of a disaster such as the collapse of the copra market, war, etc., e:penditure would first be reduced in ways which would not affect the producer. It has been calculated that a fall of 25% in the price of copra would not affect the natives' income. It is estimated that present reserve funds vould be sufficient to meet four years of slump, or other difficulties, while guaranteeing 10 pounds per ton of copra to the producer. The size of these reserve funds, which are continually increasing, is such that one may contemplate the possibility of an expert enquiry into the utilization of secondary products from the coconut palm. This view has encouraged us to suggest a programme for the improvement of coconut groves in the whole archipelago, in the belief that the co-operatives could meet the outlay necessary for such a programme, as outlined in our recommendations. (2) (1) The Handicrafts Society has been taken over and run by the Colony Wholesale Society since 1950. (2) It should be borne in mind that although the Reserve Funds may be large most of the Societies are still in debt and the value of their assets is extremely doubtful. The Co-operative Societies could not in present circumstances possibly meet the cost of the Coconut Palm enquiry envisaged by Dr. Catala since their funds and energies must for some time to come be devoted to consolidation and the attainment of a satisfactory financial ean i (Note by the Administration of the Gilbert and Ellice Islands Colony -156- We have observed that, apart from European "supervision" (quite in- dispensable), the whole co-operative movement is in the hands of the natives. It is astonishing that the individualistic character of the Gilbert Islander should have fitted itself so quickly into these community organizations. The beginning was certainly very difficult, but the natives themselves assured us that what had impressed them most was the sense of ownership which they rapidly acquired toward all the equipment which works before their eyes, and which they know belongs to them - from the truck which comes for their copra and which enables them to travel about the island, to the ship which takes that same copra to Tarawa and on which they themselves may travel. Their money is working and they are seeing it working. Whether or not they are aware of the security which the financial funds in the banks represent is of less importance than the fact that this security does exist. A few discontented, or perhaps ill-informed, people (not, however, in native circles) have disapproved of the fact that the oimer does not receive a little more money for the copra which he produces. But we are of the opinion that the outstanding merit of the promoters and the or- ganizers of the co-operatives was precisely this ability to impress upon the members the wisdom of forming reserve funds as a precaution against bad years, rather than squandering their money. And "squander" is the right word, if one considers the rapidity with which money can disappear in a few hours. Even if the Gilbert Islander wanted to accumulate per- sonal savings he could not do so, as he clings so tenaciously to the current practice according to which parents and friends may quickly im- poverish any native returning to the village, rich from a sale of copra or wages earned abroad. This practice of bubuti divides the money to such a degree that it leaves to each person only a tiny purchasing power, sufficient to buy insignificant trifles such as sweets, fripperies or other useless articles. So, if the native earned double what he does for his copra, his position would be exactly the same. And, as the Co-operative societies Officer told us, the more money the native has, the greater his tendency to neglect the work which provides the money and the more he up- sets the balance of his diet. Formerly, in the days when a few big firms monopolized trade, the native received more, but no reserve fund was established. It is ob- vious that the Gilbert Islander, with the little money which his individual income represents, would never have had the advantages which he enjoys today from the trucks and ships of the co-operatives, and in particular the latter, which enable him to travel from island to island in a region so lacking in means of communication, and in which the ancient knowledge of navigation has been lost. Therefore, we cannot but admire unreservedly such a remarkable or- ganization which has only some small defects which time, much better than any detailed recommendations, will remedy. CHAPTER 16 CONCLUSIONS AND RECOMMENDATIONS PART I: A PROPOSED PROGRAMME OF AGRICULTURAL IMPROVEMENT Experiment Station Any programme for the improvement of Gilbertese resources will hinge on the creation of an agricultural experiment station, which will have not only technical but educational functions. Its establishment is therefore of the utmost importance. Even within the limits of a modest and reason- able undertaking, the work of the station will be effective only if it is carried out under conditions which reproduce the customary native environ- ment. The station should be an experimental garden devoted chiefly to the improvement of economic plants already existing in the islands rather than a plant introduction garden. Along these lines, it would be excellent in providing useful and practical demonstrations. The slow response of the coconut palm to atteipts made to improve the yield precludes any hopes of spectacular and rapid results, and intense droughts will further delay them. However, the programme should be envis- aged on a long-term basis, and the two objections just mentioned should not discourage its implementation. The projected station will have to deal with many other problems be- sides those of the coconut palm. It will be necessary to study the selec- tion and improvement of species of basic economic importance: Cyrtospermna, Pandanus and Artocarpus. The station should also give attention to other plants of secondary importance, particularly the banana, fig and Tacca. Finally, it will be essential to establish nurseries for tree species in- cluding trees already present in the group but gradually disappearing in the absence of replanting and those which are recommended for introduction: Casuarina and Araucaria. Proposed Programme for the Coconut Palm At the moment, it is rather difficult to estimate even approximately the expenditure involved in the implementation of a programme of improve- ment for the coconut groves. Such a programme can have really profitable results only if sufficient means are made available. It necessitates, in addition to the organization and functioning of the projected experiment Station, the permanent presence of a coconut specialist assisted by Euro- peans and natives. Other expenses, although temporary, mustbe taken into account; for instance, a survey by a hydrologist would be very useful for the practical study of the rocky platform in some islands such as Nikunau. After much observation, we came to the conclusion that a series of cracks in this rocky platform made at regular intervals with explosives might im prove the condition of coconut palms growing above it. However, as we cannot forsee the exact results and do not know whether unpleasant reper- cussions might follow, we think that a specialist in geology and hydrology -158- could formulate a plan of blasting to be carried out on definite sites. The cost of this work would be very low, and it might then be possible to give thousands of coconut palms a chance to develop better and increase their yield in proportion. This specialist would, of course, also help in the in- provement of the culture of other plants such as babai and his role will be discussed again below. “isle Assuming that a coconut improvement program should extend over a mini- mum period of fifteen years (the necessary time to obtain tangible results), and that the expense involved should be approximately several thousand pounds per year, the invested capital could probably be amortized quite easily oe If this programme could be implemented in a constructive spirit simi- lar to that which has led to the success of the co-operatives, one could expect an extra production of copra more than sufficient to redeem in a few years the capital invested, even with prices lower than they are at present. It is understood that only unproductive or nearly unproductive palms would be replaced, so that their destruction would cause as little loss as possible. Considering the relative nature of any forecast under present conditions, and because no production data for a long enough period are available for these islands, it is impossible for us to cite even ten- tative figures. To follow another train of thought, the output of copra made from nuts on certain islands will always be very low. Thus in the southern islands where drought conditions are more intense, the number of nuts re- guired to make a ton of copra will be two or three times that in the northern or central islands; even during periods of nearly normal rainfall the difference will always be great, because of the generally more adverse conditions peculiar to the south. It is expected that general improvements and the necessary replacement programme will, nevertheless, increase the munber of nuts produced on these islands. Excessive optimism should be avoided in this connection, and we strongly advise leaving aside figures for the southern islands when compiling production forecasts, in any esti- mate established on a large scale. Even after an improvement programme, the increase in production of these islands would probably be very small. The overall increase in production which should be expected from the ten other islands should be in relation to the effort made. Technicians studying a large scale improvement programme would see whether the in- creased production of a limited number of coconut palms in each island would be sufficient after a time to amortize the expense involved. But it is obviously necessary that this amortization not be at the expense of the income to which the natives are entitled for each ton of copra pro- duced, or of the normal quota of eating and drinking nuts that they require- It seems from the data in hand that this can be avoided. On the other hand, it should be made clear that the problem should not be considered exclusively from a financial point of view. There is, indeed, -159- another important aspect, which is really the human aspect. In fact, the question here is to insure the livelihood of a regularly-increasing popula- tion, and the projected improvement program concerns the basis of Gilbertese subsistence, the coconut paln. We recognize, however, that a government faced with so many other problems more important than this, cannot tie down capital in such an in- vestment for a very long period. Another obvious solution suggests itself, that is the undertaking of this programme by the co-operative organiza- tions. Two reasons favour this proposal: the first is that these co-opera- tives build up annually - chiefly because of the high prices of copra over the past few years - large reserves in comparison with which an annual expenditure of a few thousand pounds would not be prohibitive. The second is that an increase in production is needed for the subsistence of the Gilbertese population. Therefore, even in the event of a copra price slump before complete amortization the investment would still be justified. If the Gilbertese are one day deprived of the income from high-priced copra, at least. their existence will be ensured by the production of their coco- nut palms and the situation will be improved if the coconut palms produce more. However, if copra prices decreased to a point where freight expenses would no longer be justified, this product, in addition to its place in human diets, could be successfully utilized in pig-raising, a source of income hitherto entirely neglected by the natives. Interim Attention to Coconut Groves Pending the establishment of the projected experiment station, 2 preliminary campaign of improvement could be undertaken immediately in some particularly neglected areas. Abemama and Aranuka Islands are the most typical in this connection, and it appears that Kuria, which we were not able to visit is in a similar condition. Considering the neglect in Avemana (with the exception, of course, of some well-tended areas such as the site of the Makin school group, and two plantations) and the presence of the insect pest Graeffea, it would be most advisable to start with this island. What should be done: t would be highly desirable that each owner should remove, in the plots of land belonging to him, palms undeniably unproduc- tive because of their age or defects, and trees killed by fire. He should also destroy the self-sown palms grown from fallen nuts, creating a dense tangle of useless and unproductive plants, and should suppress ruthlessly the abnormal or unhealthy specimens, some of which are breeding grounds for pests. He should make more frequent inspections of his land to collect fallen nuts. He should heap all the scattered debris around the palms. Fortunately, Oryctes rhinoceros does not exist in these islands, so these debris heaps do not entail the risks found in other regions of favoring its development. Suitable areas abandoned without reason, or devoid of palms through fire or other causes, should be replanted. Pending the day when ~160- real selections can be effected, seed nuts should be taken from palms offer- ing a combination of favourable characteristics (the natives know the best ones). This will be the last of our recommendations on what could be under- taken immediately. If further guidance were needed at this stage, useful advice could be given in relation to nurseries for seed nuts, to the bank- ing of earth around certain coconut palms under special conditions, and to various fertilizers. Errors to be avoided: To pass from the present state of affairs to the exact opposite would mean upsetting the existing balance. In other words, one should refrain during the early stages from completely clearing the ground as is done in plantations. Clearing away all elements with a fer- tilizer value such as rotting trunks, coconut husks, and old leaves, should be avoided. Shrub-like species, such as te uri, te mao, and te ren should not be systematically destroyed for the sake of tidiness; this applies even more to low plants which are often of considerable value, especially in areas where the palms do not grow very densely. Recommended restrictions; Because of this balance which should be preserved, it would be advisable not to permit the introduction of certain mammals, and in particular rabbits which have already caused untold damage in other Pacific Islands. Goats would be another undesirable introduction, but probably could not survive on their own on low islands. Quarantine measures for protection of plantations: We are firmly convinced - and we are not alone in this - that without the coconut, the native would not be able to exist for any length of. time. Without the food resources from it, chiefly the valuable toddy, the life of the Gilbertese would per- force be most tragically unbalanced. Any serious attack affecting the ex- istence of these palins would therefore threaten the very life of these people. We have seen that in the Gilberts, the coconut palms are free of the major pests found in other parts of the Pacific. However, Abemama Island is the scene of a very slow but indisputable increase of Graeffea cocophaga. Therefore, in the matter of careful protection of the coconut palm in the Gilbert Islands, equal care should be devoted to the control of pests al- ready present and of those which may be introduced later, Internal measures concern only the territory of Abemama. They should be directed constantly to the avoidance of any possible transfer of Graeffea to other islands of the archipelago, especially to the northern islands which have a heavier rainfall and would probably be more suited to the rapid multiplication of this pest. The shell of the eggs of this orthop- teron is extremely hard and resistant. The egg hatches out only several weeks after being laid. These two factors therefore increase the chances of eggs hatching out in places far from the infested area. While we may wonder why this insect has fortunately not yet spread, we must admit that this could easily happen. For example, the eggs fall from the leaves into -161- open bags during copra-cutting or gathering. The bags which are likely to leave the island should therefore be very carefully examined. External communications, while infrequent, are nevertheless quite suf- ficient to facilitate the introduction of insects which may become danger- ous pests. Cargo should therefore be very carefully checked, not only in the Tarawa area, which is the first port of call for most ships, but also in’all the islands of the archipelago to which some ships sail directly. Numerous other examples could be given of the many possibilities of contam- ination from sources outside or inside the group. In fact, to ensure ef- fective control, an elaborate quarantine organization will have to be established and we are well aware of the difficulties involved. In brief, the control of imports from overseas, including ill-considered introductions of new varieties of coconut palms to areas chosen at random (this could extend to the introduction of any plants), the institution of adequate quarantine and, if possible, of cleaning-up operations in terri- tories such as Abemama constitute the basis of this recommendation. Since it is intended to safeguard the basic wealth of the Gilberts we feel that no effort should be spared to ensure its implementation. In several cases, we have only touched upon certain questions and others have been purposely neglected because they are related to problems restricted to certain islands or strictly connected with local politics. Generally all the recommendations for the improvement of the present con- ditions in coconut groves come back to the need for better land utilization. While the Gilbertese frugally make the best of everything in their daily life, they neglect much of their land. The population of these territories is increasing and it will be neces- ary, in the interests of the people, to survey much abandoned or neglected land. There are even some islands where ancient constraints still burden the inhabitants and make them ashamed to work certain lands for fear of be- ing mistaken for slaves. Such factors are so many hindrances to the in- creased activity which the authorities would like to encourage. The pres- ent situation is far from alarming, but the problem appears greater when considered in relation to the future. It is likely, however, that an im- provement programme would easily bring results within 15 years and that the increase in production would then be sufficient to meet the increase in population. But it must be said that any improvement undertaken would, initially at least, encounter everywhere the indifference and indolence of the natives which are quite understandable. Observers in other areas of the Pacific have made the same remark in connection with the same problen. No attempt to improve the coconut groves and generally the material living conditions of the natives will be successful unless they themselves lend their support to it. The Gilbertese are not systematically opposed to some- thing new, but they are not inclined to persistent effort, especially if they have no proof that this can be quickly profitable. It is unfortunate that the coconut programme ranks first among the improvements to be carried out. The coconut palm is a slow-growing plant and the best attention given -162- to it will bear fruit only after many years. In other words, spectacular results should not be expected from the experiments. This does not mean, however, that they should not be attempted, though imuch perseverance, great consistency in the aims, and the most diversified forms of propaganda will be needed, Finally, it must not be forgotten that any result obtained on the Gilbert Islands will automatically apply to the coconut groves of other low coral islands in the Pacific where prevailing conditions are comparable. We would thus be tempted to suggest that any proposed improvement programme be planned for a wider area than the Gilbert Islands, and that governments controlling territories with the same problems consider participating fi- nancially in the improvement programme for the Gilbert Islands group. Improvement of other useful plants Cyrtosperma: -Although babai is not an absolutely essential food, it appears that the Gilbertese, who are so fond of it, eat it in such small quantities only because of the trouble of growing it. As this plant can only grow in pits, its area of cultivation is very limited and could be increased only by very hard work. We have pointed out that we never saw recently-made pits. We can understand why the native, once he has an assured minimum of babai, balks at such arduous labor which must be undertaken with tools only slightly less rudimentary than those of his ancestors. We wonder whether the physical difficulties which are such an obstacle to the expansion of babai cultivation could not be eliminated to a great extent with the use of modern machinery. Such equipment could excavate as far as the rocky platform, clear away the earth and finally break the platform. This recon- mendation is closely allied to an earlier one (see p. 158) which advocates @ preliminary survey of the problem by an hydrologist, whose advice would be essential before undertaking such an experiment. A "pilot" experiment could be tried out in one of the southern islands, e.g. nn Onotoa or Niku- nau, where an increase in babai areas is most necessary. It could then be tried on Abemama (particularly at Kena) where so many babai pits have been abandoned. We have not overlooked the main objections, the first of which is the difficulty of unloading under the landing conditions prevailing on most of these islands (the days of landing craft being past). The cost of such equipment would obviously be high, but one can imagine the satisfac- tion of the Gilbert Islander at seeing completed in a few days the work of clearing the land, digging and breaking the rock which would take him years were he to undertake it himself. It is certain that any such excavations _ would at once be planted with babai. Let us repeat that we do not advocate here the general use of modern machinery all over the islands but merely an experimental operation. Colocasia (taro): Our study of the Colocasia showed the advantage of using old abandoned babai pits. The native must decide if this work of reclama- tion is worth while or not. Pandanus: It has been pointed out that because of high copra prices, sev- eral islands showed an increasing tendency to neglect the pandanus trees -163- which are replaced by the coconut palm. It is absolutely essential that the present stands should be maintained, and it would be advisable to in- crease them as a precaution against the uncertain future. Artocarpus: The breadfruit tree is everywhere well tended. Any future agricultural station, however, will have to improve the seedless varieties, in particular. Fig trees: The distribution of saplings in the villages could be studied in order to increase the present insignificant number of trees. They would provide an excellent additional food for children. Bananas: It has been noted that banana plants were very rare, except in certain centres where serious efforts have been made - missions, schools, etc. The natives, the children especially, are so fond of bananas that these should be planted in the neighbourhood of all villages where there are old abandoned babai pits already more or less filled in. Dwarf varieties should be tried because of their ability to resist winds, and their large number of shoots. Like Stone, we would advise cor- recting iron deficiencies by the application of soluble iron, and nitrogen deficiencies by covering the soil with a thick layer of leaves and organic matter. Papaya trees: It is recommended not only to increase the number of papaya trees in the villages, but also to introduce varieties likely to appeai more to the adult taste. Tacca: The advantages of Tacca, so fully described earlier, are such that its expansion is recommended; this food can be grown between coconut palms. Instruction in improved methods of eliminating the toxic element and in less primitive techniques of extracting the starch will be needed. Tobacco: This plant is successfully grown even in the villages of the southern islands, where the natives, in the absence of twist tobacco or of money to buy it, fall back on the few plants which grow round the huts. It is recommended that an expert should teach the Gilbertese how to prepare the leaves correctly in order to improve their smoking tobacco which, though still not of good quality, would be better than the pitiful product which they now obtain. Other plants: We have listed earlier the plants the wood of which is used by the Gilbertese, but this use is too often restricted by depletion of these species. In spite of the slow growth of some of them, we recommend replanting. Hach village could well replant the surrounding areas, to some degree, with a few useful trees. Our preference would be for te itai (Calophyllum) and te kanawa (Cordia). Native preference will obviously be for trees with multiple uses like te mai (Artocarpus). This species, how- ever, has to be cultivated. Found today only in villages where little space is available for increased plantings, its cultivation could be at- tempted in inland areas where ground water and soil conditions are favourable. ~16). The main experiment which should be tried, however, concerns two trees the introduction of which, if successful, would have the happiest results. These are Casuarina equisetifolia and Araucaria cookii. From a very fine specimen on Butaritari, we know that Casuarina can grow just as successfully in the Gilbert Islands as on other coral sands. Areas near the sea at the ends of islands and islets are to be recommended. Araucaria is found in very dense stands on many coral islets in New Caledonia where edaphic conditions are similar to, and average rainfall not substantially different from, those of the Gilbert Islands in normal times. Obviously, the resistance of these Araucaria to very long droughts cannot be foreseen, but they are worth trying along the coastal areas, preferably on the lagoon side and in some central regions unsuited to the coconut paln. Finally we should like to repeat this essential recommendation: there should be no decrease in the number of plants of the various species which play a primary role in the vegetative balance necessary for the flourish- ing of plants of high economic value, such as the coconut palm. The coconut groves may be "tidied" without destruction of the useful shrub species. It must always be remembered that undergrowth is of great importance for the production of humus and the limitation of evaporation. PART II: EXPLOITATION OF MARINE RESOURCES Basic marine resources It has been seen in Chapter 11 that in the Gilbert Islands fishing is a daily activity, whether outside the reef or in the lagoon, or limited to the collecting of a great variety of organisms on reef flats uncovered at low tide. We indicated that if the natives so wished, and if the call of the sea attracted as many young men as in the past, they could easily increase their resources in sea products. We also pointed out that the number of canoes is small in proportion to the density of the population. We would, therefore, recommend that the Gilbertese be given all possible facilities for increasing their equipment (canoes and fishing gear), and that these facilities be supported by PROS conducted propaganda and even with a certain amount of subsidy. It should be stressed, however, that fishing should remain a native affair. Anything of an industrial nature would entail the risk of most unfortunate repercussions on the fish populations around these islands. In all our conversations with old fishermen we could sense that the balance is very precarious. These men of great experience in sea matters were of course unable to cite specific facts, but they had an intuitive knowledge that any increase in the quantities of fish caught would destroy the balance which enables them to count upon sufficient daily supplies. We are, therefore, obliged to recommend that no attempt should be made to in- tensify, for example, the fishing of large tuna (te baibo) in the northern and central islands. -165- An exception, however, might be made in favour of the southern islands. It has been noted that in periods of under-nourishment on these islands, during droughts when land crops become less plentiful, the Gilbertese fish less often because they are disinclined to make the necessary physical ef- fort. One of the co-operative boats might be fitted with simple gear which, in times of restricted food supplies, would make it possible to catch a substantial amount of fish which would be a valuable additional supply. This boat would fish for a few days around each island. If the fish were not plentiful enough in the waters of the southern islands to justify such an undertaking, nothing could prevent the northern islands from supplying the required quantities during the tuna and bonito seasons. The native crew permanently attached to the boat could be reinforced with extra fishermen chosen from among the best on each island, whose ser- vices would be required only while the boat was in their own territory. There would be no need to bring European personnel from outside, with the possible exception of a specialist in the gear to be used, whose presence would be required only for a time. We met several boat masters whose curi- osity in sea matters and outstanding knowledge of the hydrography of the most difficult areas fit them well to carry out this type of operation. However, in the field of protection of fish resources, one question seems much more urgent, and should, forthwith, engage the attention of the authorities. Elsey (1951) writes: "There is no doubt that the Japanese, with a population increasing at more than a million a year, are scraping the bottom of the barrel so far as fishing is concerned, and the situation is one that cannot easily be dealt with on a satisfactory basis." We also find (Anon., 1950) a short article which states: "Japanese Fisheries Board officials hope that after a peace treaty is signed they will be al- lowed to fish as far south as 5 degrees South of the equator, BCON (Army newspaper) said recently. At present Japanese fishing is not allowed south of 24 degrees North of the equator." /weitten in 19h97 In view of our previous comments, our fears that the Japanese fisher- men may come and "scrape the bottom of the Gilbertese barrel" will be widely shared, all the more so for their already having an intimate know- ledge of the waters in these regions. . The consequences of such an intru- sion would be disastrous. Hence our recommendation that there should be established forthwith around the sixteen islands very wide territorial zones, strictly closed to all non-Gilbertese fishing boats, except with special authorization granted only after the most careful examination. The areas traversed by migrating tuna, bonito and other fish should also be protected. Secondary marine resources We have seen that the number of profitable marine products is extremely limited; that the native sponges are not of sufficiently high quality to reach a wide market; that mother of pearl products are either non-existent or insignificant; and that good species of holothurians are probably not -166- sufficiently numerous to justify development and do not interest the Gil- bertese. There remains, therefore, the only product which actually has a commercial value: shark fins. In spite of the small extra income brought in by this product, we do not think that it would be possible to persuade the Gilbertese to increase shark fishing deliberately. We can only recommend better methods of preparation and a study of the best pro- cedure for stocking, pending the reopening of the Chinese market. Sponges: It might be useful to examine the possibility of introducing, in certain islands of the group, sponge species of high commercial value such as Spongia officinalis subsp. molissima. This sponge was successfully cul- tivated on Ailinglapalap in the Marshalls by the Japanese (Tressler and Lemon, 1951; p. 739), who asserted that it is native in the Marshall Is- lands. But it is debateable whether in competition with existing species of negligible value, the latter might not win. Chanos and fish-ponds We would recommend a better use of certain fish ponds, especially of small inland lakes like those of the island of Nikunau which, because of their size and location, offer ideal conditions for growing baneawa (chanos). PART III: MISCELLANEOUS RECOMMENDATIONS Handicrafts The first condition for the success of those handicrafts most likely to have a regular market in considerable quantities is that they should be uniform: the native must be convinced of this. For example, table mats would sell readily in overseas countries but only in matched sets. Un- matched, their appeal would be limited. A disciplined system of working should also be established. Other defects that could be corrected are usually the result of ignorance of commercial psychology. Many Gilbertese girls have learnt in the missions to do embroidery work, cross-stitch, etc. which is very fine and reflects as much credit on the teachers as on their pupils. But the designs selected are generally European while those with a South Seas motif would undoubtedly have stronger sales appeal overseas. There is no lack of subject - huts, canoes, fish, etc. The natives them- selves took a surprising interest in the designs shown in Plate XVIId. It is also necessary to find markets. This will be impossible, how- ever, if sample consignments are not uniform, and if the purchasing firms are not certain of receiving articles strictly in accordance with their orders. Possibilities of selling abroad, especially well-finished arti- cles, can be examined but will have to take into account the prevailing fashions and the competition of similar products of equal and even superior quality from China, Japan, etc. Finally, we were told that one of the most likely markets would be Australia but that its customs tariff was too high. The small amount of handicraft produced in the Gilbert Islands would not -167- upset the Australian balance of trade, and considering the help this would be to the Gilbertese, it is most desirable that these excessively rigid conditions should be modified. There was even a hope that the South Pa- cific Commission might lend its support in this direction. Pigs In the course of our brief study of the diet of the Gilbert Islanders, we have pointed out how little pork they eat, and how their lack of inter- est in pig rearing is simply a result of extreme indolence which we do not think can be overcome by any amount of advice. However, the problem of pig raising should be considered from an angle other than that of supplementary food which, for the moment, the Gilbertese do not really appear to need. Provision has to be made for the day, which we hope may be as far off as possible, when there might be such a slump in copra that its transport would not be justified. Ii that day were to ar- rive, the money available to the Gilbertese would consist only of the re- serves accumulated by the co-operative societies during prosperous years. However large these reserves, it would be essential to make them last as long as possible. Every device that would help in achieving this end should be most carefully studied. The Gilbertese would have very little money, but more copra than he knows what to do with. It is then that pig raising would be an important resource. The question may be looked at in this way. There are two phosphate extraction centres less than two days distant by steamer which provide ready-made markets, and their buying possibilities are relatively good. The British Phosphate Commissioners must provide the entire food supply for a certain minimum number of people. It is of course to the advantage of the Commissioners to use the empty holds of their ships to import from Australia the necessary supplies for the entire population of these is- jands which, apart from a little fresh fish and a few chickens, produce nothing but phosphates. Without doubt, should copra prices slump to an unprofitable point, the British Phosphate Commissioners would not refuse to assist the Gilbert territories by buying their pork production, especially as the majority of the workers who extract the phosphates are Gilbertese. The volume of Australian trade would not be affected by this preference, while the assistance given to these people might be of real importance. From the technical and practical point of view no great difficulty is apparent. Gilbert Islands pork is excellent in spite of the sparse, unvaried feeding. An improvement in these animals might immediately be attempted on a very small scale; later the introduction of selected breed- ing animals could be considered. Should breeding be thus organized, the pigs should receive more abundant, and, if possible, more varied food than they do now. In addition to the various plants which are sometimes given them (see Chapter 8), nothing would preclude the addition of a certain amount of fish to their ration. Since the Gilbertese would have no need to limit their personal consumption of coconuts, they might perhaps be -168- induced to divert for the benefit of the pigs a few of the ikari which every month fill the fish traps, often in thousands. On the islands where this species of Albula is rare or non-existent, groups of fishermen could always find enough Serranidae on the edge of the reef, or bonitos further out, to make up for the absence of Albula. Finally, expansion of Tacca cultivation, the only food plant which can grow under the coconut palis, could probably provide an additional source of food for pigs. Pigs could be loaded alive onto one of the co-operative boats (which would have little else to transport) and brought to the slaughter house which could very easily be located on Betio (Tarawa). There, after cutting up, the pork could be properly salted so that it could be taken in an ex- cellent state of preservation to Ocean Island, thirty-three hours sailing from Tarawa (Nauru is a few hours further away). It could then be kept in cold storage. If we deal only with the population of Ocean Island, approximately 2,000 people, and if we count only three rations of 180 gers. of pork per individual per week, the warket offered by this island would be fifty tons, which represents a stock of 1,000 pigs. If the Gilbertese would take the trouble, and if the British Phosphate Commissioners would agree to pur- chase this production, then breeders would enjoy substantial additional in- come. Diet It it recommended that a dietitian, already familiar with native food problems, should make a lengthy stay on the Gilbert Islands. He should be of an age and physical condition permitting him to dispense with modern comforts in order to be in close contact with the islanders, and to move from place to place quickly and easily. As well as his research work in the field, this specialist should arrange for further analyses of specif- ically Gilbertese products to be carried out by the laboratories best equipped for the task. Additionally, he could give the authorities the necessary guidance to enable them to curb the excessive consumption of certain products and increase that of others. Among many other things, this specialist would have to study the vitamin B value of brewers yeast extracts (such as Vegemite) and specify the amount of these extracts to be used for a given quantity of rice and flour, in other words, what added amount of vitamin B is sufficient for the complete utilization of the carbo-hydrates. The work of this dietitian and the information he collected covld form the basis of a practical dietetic plan, applicable to other low coral is- lands. General Conclusions in the form of Recommendations We do not wish to disguise the unavoidably limited scope of many of our recommendations; we are aware that it is easier to make than to ) =169- implement them. We have, however, attempted to consider them all from a purely objective viewpoint, limiting ourselves to the strictly economic aspect of the survey entrusted to us. We have repeatedly emphasized the importance of the oft-threatened equilibrium between the people and their environment and the need to do everything possible to maintain it, both by preserving vegetable and ani- mal sources of supply and by watching the diet of the people in the Archipelago. However, if we had had to study the problem as a whole, we should have dealt with additional aspects, for instance the need for a type of clothing suited to the climate and the importance of a cultural equilibrium, an essential condition for a happy living. It was with plea- sure that we noted that the local Administration was also imbued with this idea. Biles GILBERTESE VOCABULARY This short vocabulary is but a provisional and very incomplete list. Some words, not appearing in Gilbertese dictionaries, may not be accur- ately spelled because our Gilbertese informants occasionally had diffi- culty in spelling them correctly. It must also be pointed out that terms used may differ greatly from one island to another. The article-like "te" has been omitted throughout. ai aine al onauti aitoa amaere amakai . amerika amori ana anabanaba aneang ang ang anga ango anibanakoi anikatouea anikomri anikomri aniwaentang anoi anrairaki antemai antibitia antinakarewe antinamainuku aoua arabaikiaro aramaiki aramaru aramaoia aranaonimali arantebwe ararikitoa arataitara coconut crab - Birgus latro female daylight catching offlying fish by trolling, using a special line fitted with floats tree, Lumnitzera littorea (Jack) Voigt stage in the development of coconut stage in the development of coconut plant of recent introduction ascribed to American soldiers,e.g. Pluchea odorata Cass. fish of genus Gerres fish of genus Hemiramphus variety of Pandanus, on Tarawa (native of Ocean Island) canoe mast airplant, Bryophyllum pinnatum (Lam.) Kurz shellfish Of genus Cypraecassis stage of growth of Cyrtosperma tuber tree, Premna obtusifolia R. Br. variety of Pandanus, on Tarawa variety of Pandanus, on Tarawa variety of Pandanus, on Tarawa - shellfish, Conus lividus L. - variety of Pandanus, on Tarawa hammerhead shark, Genus Sphyrna variety of Cyrtosperma (also called te babue) inedible inner part of Artocarpus fruit variety of Pandanus, on Tarawa variety of Pandanus, considered best of all variety of Cyrtosperma on north islands (Marakei ) four o'clock, Mirabilis jalapa L. one of the favourite varieties of Pandanus for preparing tuae variety of Pandanus, on Tarawa variety of Pandanus on south islands (Nikunau) variety of Pandanus whose drupes irritate the tongue; this is also one of the favourite vari- eties of tuae variety of Pandanus, on Tarawa one of the favourite varieties of Pandanus for preparing te tuae variety of Pandanus on south islands (Nikunau) one of the favourite varieties of Pandanus for tuae Lalas aratekura arateman aratokotoko ari arinai aronga ata ati atine atinimainuku atuaro aua auamaran auatara aubunga awal awaneari awateai ba baara babai babue baibai baibo baiku baireati baitari bakoa baku bam (= palm) banaba banana baneawa bani banikaina ba-n-te ni banu bareki bari baru batano batua batua one of the favourite varieties of Pandanus for tuae variety of Pandanus, on Tarawa variety of Pandanus, on Tarawa spathe of coconut palm. Used also for the bound inflorescence from which toddy is obtained fish of genus Scarus shrub, Acalypha amentacea var. rope from mast to outriggers of canoe bonito. Pelamys sp. small tuna. WNeothunnus sp. variety of Cyrtosperma on Marakei small bonito fish of Mugilidae family, in general fish of genus Mugil fish of genus Mugil large-sized Tridacna fish, Aprion virescens Val. variety of Pandanus on southern islands (Nikunau) fish, large specimen of Chanos chanos Forsk. leaf fish of genus Scomberomerus (Tasard) Cyrtosperma chamissonis (Schott) Merr. variety of Cyrtosperma on south and centre FSTAS 4 44 ane Lee fish of the Pleuronectes group red-fleshed tuna Neothunnus macropterus Schlegel (yellow-fin tuna) fish: rays in general tree, Barringtonia asiatica (L.) Kurz edible jellyfish of genus Tamoya, Carybdeidae family sharks, in general top of rhizome and base of petioles of a Cyrtosperma used for artificial proliferation of shoots a cultivated palm Ocean Island banana, Musa sp. fish, Chanos chanos Forsk. eoconutilear™ 1s) Pandanus leaf coconut palm frond fallen coconuts eaten by rats weed, Physalis angulata L. (southern islands) fish of Carangidae family fish of Serranidae family "rocky platform" of medium hardness fish of genus Canthigaster small clam of genus Hippopus =V72= baua baukin (= pumpkin) baweina bawemara bebe bekei ben beneka (= vinegar) benu bero biku’(= re a=) "a TnIspes.tq poe *edTai ysoute 4IM717 jo AjatreA UT EU JU Sueey ay, °p ‘q[nIjpeetg jo AVatzen Tew eh hae) *syunI2. snuepueg peyouerq jo *punorzs uo peerds ‘SutArmp jo epem STeETTTd °peus Sutyoop °-q a3e1S 1SBT UT 9Bn 24 JO soyen °e ‘sTeuTue pue UsIPTTYO WoT ButTpees *ssutyqino Aq peonporda.s Sunofk Sut109901d sqieu JeatT ynNUuOD0D °p forertyng 04 AYeTIBA sseTpesg °9 rete ae? z ie “] pe are = Ps * hg 9 Leal 4 tur Cea OUVMINNG TL Suen Lat “uTYeU TU Suey 34 °eYyTTeZTeuU 34 *uedo 1nd YIM SsaetTJeTITeA owes °q *‘Teul 54 SOTYSOTIGA JO JBoT pue YIN eS d. a. Te mai breadfruit, 4-5 years old, with frame of coconut trunks holding compost. Some branches withered by drought. Banana plants in old babai pit. Pig enclosure. Carrying compost for babai plants in a coconut leaf basket. Plate XIII. Ce b. Babai plants, about 5 years old, well-tended. Baskets holding compost around babai plants. b. Pumpkin plants in village on Nikunau. a. Vegetation at top of beach. e. Tobacco plants in village d. Rnizophora mucronata on | on Nixunau. Aranuka Island. | Plate XIV. “AX 922Td °> AX UT ; umoys dei}z ut JUusTU suo ut pedde1z TIBYT OOOSE JO YoVeO B JO YTeG “Pp "Id “STI YOVSHS 98S “TOY eMezey, JO epts ues00 uo *derm-ysty euoqs °9 ‘uns 94} UT “TTOVy BMerey, suTkIp TreytTeq 04 Jo yred aTqtTpy °q "TreztTeq 94 YysTs ATTef eTqtpa -e ay Ne As ‘iby ly “4 ‘ums 9} UT Taeyt JO syeTTs BuThzq °e a “d . He iq *TIAX 29% ; z aod °oT1aN4 UeeIH °d *sauaos 9Sa7.IEqTT) &q pertdsut sustsep AZeptoiquy °p 2s (Ser "der, Ted °4 “SULIOJZETA SuTystTy se pasn sautyeuios *sagnoyyno eaT7eN °B ATOLL RESEARCH BULLETIN No. 60 Climate and Meteorology of the Gilbert Islands | by Marie-Héléne Sachet Issued by THE PACIFIC SCIENCE BOARD National Academy of Sciences--National Research Council Washington, D. C. October 31, 1957. Climate and Meteorology of the Gilbert Islands 2/ by Marie-Héléne Sachet The climate of the Gilbert Islands is an oceanic tropical one, devoid of great fluctuations in temperature and almost without severe storms. There is a gradient from high rainfall to low, from north to south; rainfall is seasonal and sporadic. Recorded data are quite inadequate for any satis- factory statistical or long-term considerations. A summary of probable weather conditions, with some meteorological information assembled in tabular form, is presented below. Accurate, dated observations to augment this record are greatly to be desired, Atmospheric pressure All of the Gilbert Islands lie close to the Equator, and are within the equatorial belt of low pressures; there is very little seasonal change, and diurnal changes, although regular, are also of small magnitude. Wind direction Prevailing winds are easterly, their direction varying slightly with seasons: they blow from NE and ENE from Janzary to April, then shift to E to SE during the northern summer; from October to January, they revert to E, ENE and NE; this is true for all the islands, although the occurrence of ESE and SE winds is slightly greater in the southern islands. This pre- vailing wind is the NE trade wind. In northern winter, N and W winds occur occasionally, as an extension of the Australian monsoon; this is more frequent over the southern, than the northern islands. Westerly winds can occur occasionally in the other months, and are very variable. Owing to the lack of large land areas, there are really no such things as land or sea breezes over the Gilberts, and winds vary little in direction during the day. Wind strength Winds are usually light or moderate, the strongest winds occurring mostly during the northern winter: see table A. The irregular westerly winds are often accompanied by squalls, bringing heavy rains. Calms are not very frequent; the Gilberts lie well to the west of the Doldrums. i/ This summary of available data on the weather and climate of the Gilbert Islands was prepared for the use of the Onotoa Expedition in 1951 but was never given general distribution. Since rainfall tables for only a short period of years were provided in the original manuscript of the report by Dr. Catala it was thought best to bring the 1951 compilation up to date and issue it at this time for the benefit of those who are interested in the Gilbert Islands. -- Ed. ae Storms and Cyclones There is no record of thunderstorms in the southern islands, and table B shows that they are rare in the northern islands. "Only one hurricane is definitely known to have occurred. This damag- ing storm crossed the Gilbert Islands on December 4-5, 1927, where it was observed in the extraordinarily low latitude of aN at longitude 17308: On Makin Island severe southeasterly gales lasted for nearly 12 hours” (Weather summary for Naval Air Pilot). The British Colonial Report (no. 1460, 1929) dates this storm January 1928, and says "the islands of Butari- tari f= Makin _/ and Little Makin were visited by a hurricane which did considerable damage to property and coconut trees, but there was no loss of ore Air temperature on land Temperature varies very little throughout the year, and the daily vari- ation is only of a very few degrees; the mean temperature is about the same for all the islands, very slightly higher for the southern ones. Air temperature at sea This is also very even, with very little daily or seasonal variation; the mean is very slightly higher in the southern islands than in the northern ones, and only very slightly lower than the average temperature on land. Sea water temperature The temperature of ocean surface water, judging by the brief records of the "Neptun" (Prager, 1903), is almost constant throughout the year, with lowest temperatures from February to May; the yearly average is almost the same as the average for air temperature at sea. Visibility The sky in the Gilberts is usually characterized by scattered or broken clouds; fog is unknown; haze is rare (see table A). "The days with 2 tenths or less cloudiness are, however, very rare on the average, al- though deep blue skies and brilliant sunshine are a particular feature of the Gilbert Islands climate..,.. During monsoon weather .... in open weather, the cristallinity of the air among the islands has received specific mention..." (Weather Summary for Naval Air Pilot). See tables A and B. Vapor pressure and relative humidity What little data are available are so inadequate that no generalization is possible, beyond the fact that these factors vary from island to island; information is very much needed. =D Rainfall Rainfall varies greatly from island to island, being highest in the northern Gilberts, and diminishing steadily southward, to reach a minimum in the southernmost Gilberts and northernmost Ellice Islands. Further south, it starts increasing again. Generally, the driest months are October, then November and September; but yearly totals are quite variable, and according to the British Colonial Reports, many years differ from “normal conditions"; periods of drought are not infrequent; here are some notes on them from the British Colonial Re- ports. Rept. 1369 (1928): "The 14 Gilbert Islands lying between 2° North and 3° South latitude are visited by periodical droughts, of which the cycle appears to be five to seven years. Butaritari and Little Makin, about 3° North of the Equator are unaffected by this condition.... during droughts on central and southern Gilbert Islands, it faverage yearly rainfall/ is sometimes as low as 3 inches. "Drought conditions threatened during the early part of the period 1924-25, and towards March 1925, were fully established. No rain fell on any part of the 14 affected islands between April and December 1925. The drought then broke, and the coconuts, which had ceased to bear, put out fresh spathes of blossom. When the spathes opened, however, rains fell with such torrential force that the young blossom was beaten from the stalks, and though the trees throve, the first coconut crop was a failure. The result was a drop of about 1500 tons in the amount of copra ne eae that year from the Gilbert Islands." Rept. 1387 (1928) "Drought conditions prevailed on the Gilbert group for the whole of the period 1926-27" (See figures for that year for Tarawa, extracted from this report, in table D). Rept. 1460 (1929): "The drought which prevailed throughout the Gilbert group during 1926 and 1927 passed away, and from November 1927, to February 1928, the group received most beneficial rains. The strong westerly gales which may be expected during that same period of the year were exceptionally severe and seriously interfered with shipping." Rept. 1879 (1939): “The rainfall during the year [i9307 was below average, and at conditions were experienced in the central and southern Gilbert Islands. Blue Book for 1940 and 1941 (issued 1944): "Exceptionally heavy rainfall fin 19h07, following severe drought of 1938 and 1939. Trade wind almost absent: frequent westerly squalls during the year, and violent westerly gales in November and December in the Gilberts.” Rept. for 1952 and 1953: "During 1952 the effects of a drought in 1950 were wearing off, and by the end of 1953 the increased rainfall of the previous two years was being reflected in the higher tonnage of copra being produced." Rept. for 1954 and 1955: “While in 1953 copra production was good, a reflection of the good rainfall in 1952 and 1953, 1954 saw the beginning of a drought in the Colony which increased in severity in 1955. Copra produc- tion has not been severely hampered, but 1956 production will undoubtedly be low, owing to the fifteen months or so which it takes for rainfall, or its lack, to affect production. Towards the end of 1955 production was falling, and the people in the Southern Gilbert and Phoenix Islands were again storing nuts for food, rather than turning them in copra." The same report elsewhere mentions that the drought appeared likely to continue into 1956. Figures for that year, or for the current one, are not available. Sources for text and tables 1. U. S. Hydrographic Office Weather summary for H.0. 272. Naval Air Pilot. Southwest Pacific. 1-140, Washington, D. C., 1943. 2. Great Britain Colonial Office Colonial Reports: Annual. 3. Gilbert and Ellice Islands Colony Blue Books. 1931-1941, Suva, 1933-1944. 4, Heidke, P. leteorologische Beobachtungen .... Mitt. deutsch. Schutzgeb. 30: 3-170, 1917. 5. Prager, M. Zur Meteorologie der Gilbert-Inseln .... Ann. Hydrogr. 31: 348-354, 388-395, 1903. 6. Great Britain, Hydrographic Dept., Admiralty Pacific Islands Pilot. 2; 1943. 7. Gilbert and Ellice Islands Colony Medical and sanitary reports for the year... 8. Cloud, P. E.,-Jdr. Preliminary report on geology and marine environment of Onotoa Atoll, Gilbert Islands. Atoll Res. Bull. 12: 1-73, 1952. ae Table A Area is) 085/59 N, 170°- 175° E Meee an Oe=t50 S,. 17O°- 175° E Meco ee) Oos'5° S, lfse-0100> E Units, or ways of Type of data tabulating data Area 1 Area 2 Area 3 Distribution of % of total wind ob- wind forces for servations recording: February calms GO Sok Mt Beaufort! 1 to 3 incl. 50 35 71 Beaufort 4 to 6 incl. 30 61 22 Same, for May calns “ee eee 5 Beaufort 1 to 3 15 76 61 Beaufort 1 to 5 25 12 36 Same, for July calus coeaah, O) Talat) * 4d Beaufort 1 to 3 60 25 19 Beaufort 4 to 6 Ke) 15 1 Same, for Oct. calms meee) cde) mo Beaufort 1 to 3 88 67 67 Beaufort 4} to 6 12 33 29 Same, for Dec. calms 13 6) am AG) Beaufort 1 to 3 65 69 50 Beaufort 4 to 6 19 31 hi5 Beaufort 7-8 3 fe) 5 Factors affecting? % of observations with: visibility for lower type clouds % 138 * 7h * 73 February exceptional visibility 0.0 0.0 0.0 haze 0.0 0.0 Ox Same, for May lower type clouds * 60 * 100 76 exceptional visibility 0.0 Taba 4.8 haze 0.0 0.0 0.0 Same, for July lower type clouds EEN BO ut 65 exceptional visibility 0.0 0.0 OES) haze 0.0 0.0 0.0 | Same, for Oct. lower type clouds * 39 * 100 52 | exceptional visibility Abie d 50.0 8.0 | haze 0.0 O. 0.0 | Same, for Dec. lower type clouds 35 86 84 i exceptional visibility aa 0.0 8.0 haze 0.0 Bnet 0.0 * Indicates less than 25 observations for the month in this area. Average velocity: Beaufort O-3 = O to 10_knots " ly = 1d to miGwenets 2 No observations recorded fog Source: adapted from l. *—T wory peydepy somos aXou IO YOUT TO’ UTM skeq /T QWaIYXG SORE TS70L ueay e) Cae SOE ere) zTaquesaq O°T Out 9°9 19°9 TOQUSAOL{ 7) 0°g co OT |c°9 1390700 0 Cae Led Git Jaquez.dasg O G*2 OG Qu qysnsny 0 °€ g°9 19°S Aqap €°0 G*h 6°9 19°S aune (0) Vivd ON G°e 9°9 JS°S Keay 0) 6°S T°TTI9°9 Trady 0) G°9 0°g 10°9 Wore o) oe 0°79 je°L Arenige,t 0 slaty 9°eT1e-2 Arenuel LO-SO6T Susteqy SUWSIVXY Te{OL et Stele 6 9 gL 06 49 ueoW 0 ;O}] OO}; OF} * fo) O*IT| 0°9 gL 68 48 Jaqusseq ) Teal eece| oleGe ) @PE) |) eS Pay 94 68 HQ Jaquicroyy O°T Eh wi} -O i) et 1 OPS | WO Gry 61 €6 98 1I3q0100 ) OF /2051- 20) dr 2 G°6 1 9°S Boe 6L 26 SQ Jaque1des 0 eta OF ete 0 0°02] Q°TT|6° el 06 42 asneny G°O Le @ i we |} ete 9 O°6T! 9°g 1S°S Ql 06 48 AToe Owe © |G |) Sai) Che i O0°9T! 9°9 pe L 06 49 oun Bae) @ || © || © G°OT] 9°9 JE°S 06 16 Kew ) @ NS 9b) tee THE ae QO Ith't gL 68 HQ Trady ) OnE On eG 9 G°ST}] 9°S 1Q°H Ole 6c eC Yyoreyl Te T Oo} O T 9 O°9T| S°9 10°9 ea); 19. So Arenige7d 0) fe | © |) © 0 OPGic|| See |eIAs gl 68 9 Arenuep HO-EO6T ; Tre, Tseqng oo oF P S | M 6. te Se ad oa P: zepunyy ak 5e| (queored ) UTM Sskep jo - WOLZ SUCTJeATSSgO jo % —c | “karprumy | “ON eBereay pUuTM Trejurey SATILBTOY “do due, ITV @ STqeL eB A a te IR RE A a POT Table C Ampeg ie Oe ise MiwliOo = 175° B Brea yee OF ia 50 Sit 7O> = 175° Area Be Ooun So SG, ioe = 160° F Type of data Units, or ways of tabulat- Area Area Area ing data ali a 3 Frequency of various 7 % of observations recording: precipitations for rain and showers 0.0 34.8 133 February thunderstorms 0.0 0.0 6.7% : squalls 0.0 entey | G7 Same, for May rain and showers - lata 9.5 thunderstorms - 0.0 Ten squalls - 0.0 21.4 Same, for July rain and showers 0.0 O.On, .) Bone thunderstorms 0.0 0.0 Tet squalls 30.0 25.0 nn) Same, for Oct. rain and showers 222 50.0 L..0 thunderstorms ORO) 0.0 8.0 squalls 5.6 0.0 LO Same, for Dec. rain and showers 10.8 ee ie){6) thunderstorms 0.0 3.4 4,0 squalls 5.4 0.0 16.0 Source: Adapted from 1. TO coo Tome — 1 tr oe eh ee enerey, Ot 6T 66° tL 6 HT ao gto C - JeeOT- 4 Ol. Ge ce “CL *e eS OT G4 eh = eg" O- TT"TT «OO"R ene rey, : 626T 00 Com aco 000 coy 1 C°O §€60 927% ELE G6°T] 6£12 96% Le‘92 Lo'LT enerey, | G26T Go wie och Gi 66.6 00-0. Oree “6'e che 05s G60} 0S*0 lg. “£91. e"0 eNerIeL, L26T jG | ee Se6) 00°O 12°0 00°¢ feed CMeTEy, 9z6T 66 = eG suelteqy 606T go°h = c9°0 Oe°t= sig suereqy QO06T LG Gree 10-0) = solo OOO == 44-0 ive Gory — Ye Ge GeO q GS'e GB20-- 98° 60°F) suereqy LO6T HeaoG= O01 = Oc Gr GOO. ete = 960) 6010 Gre core 99'G ETS O)0 Ol te suereqy 906T Si On ae eae Se OCG) icon Oe cl cole 20 Ola Oc ce 16 0c Toh “ho HE sueleqy . GO6T Gl°6 Ors sueteqy C/O. 1G Eee Ceccnl Oceye «OGe a Tore —6c°O. 980-9) feo Tapyrzanng HO6T Co, Tosa CO O = Oe ee oe tf — OL SEG OOrie Gli O6°GT Gh=S Trey Treqng COT T#Q0L ‘99d "AON «= *900 «6 *9deg) «= ‘Suny = Atne-— aun Aen [tady yoreq ‘qaq” ‘uer puetst Ssoyour UT TTesurey ATYWUOH SpUBTSI J1EQTTN oy} AOZ SatTqey_ TTesuTey ad 9TqeL 6£°TL G9°T6 29°66 oL°09 Teq04, eoooqoo0o90 e e . ° e . . OMNNOWONDODODOSO ADNAA-OOCOOOOMO0OS +T+OOOTnHOOOOOCCO0O090 “dog AMODOAHDAON QYUN iG) (S/S) AUST aAOFeARNOS MA WN ON MFOO0D90900COFAHOAOCOOO * AON MAUVHSOOSSOSCOMSSSOOS OoAMOO 9e000 O00 OF 9 CLAMS OSSOSOORSOOAHOS "4.00 NFUSCKCOCOCOHONONMHNA ° OMOrFeOKOmM (A ae Los rTssSbSSgoggadraa en Nest Ast WN QkARSSSanms ONVNMNAMANANMA MA A SO O25; S61 _ FO% wore Gort LO RT SSE feet Ste Goal. West Car Iyer osc }86 EH Gear mCuy TMQr O63 YO Jae Eye. —SS6°Or OnE Seis0. 36657 HO-e sO STGEOT GL°h €6°OT g0°6 Hos. SOSG-- eTZ<6 £6°0 9g°o UTS SIeee Gee Oe0 Son tts <2 Sit “ony AjTne oune "p,quoo *q oTgeL 9 °2 Oe LenS Gy T Gye’ 6L°6 9c°OT 0S*2 uice as 66° ST GS°9 ong €y°€T GO°eT aeleeds ke oa Bid Hi 6 GH GE°g 6L°€ 98°9 ES aELE 06°2T FAG €q° LT GT°9 otenddoodHa tO Mh NOW Nal NOOO MT AAO ae AO Sisie skGer Colne stQs9E UO +S aye ey 96°8 92°9 Ttady yorey e ° NOOt-OMO NUN A OWMIA 2) eh CO@ Coo “qe OVE eonez,Tqet 00°O TINOUON OT" oT * eynuedly €9°9T eT any ae 6) eueule qy 00°0O BUBTEIN teed eMe IBY, Gh suereqy £G°0OT ToYesrey eo°eT) = =—«_« Fee zeqng O9°ST ULMEN 9TI4TT €€6T SseIoIy euemey, 203.0U0 NeUNHE N n2ieg eonayTgeiy, TINOUON eynuely eTINY eureUuls qy BUCTEN CMC ICY, qi’? SuUeTeQY 6h°9 Toyesrell c9°9 Trey Treqng T9°H UPTON STITT cLOL sueteqy TOyereyl TICLE Te Ng UTYeEN STITT TE6T “uer pueTst eq-Te 0020, 0020, 0070. 00:0 -00°0, 0070 69'0 (eg = COs0l- idee cepOn Gone, “UerO -tUrce Gc 20l0s _00eG- Goren Ooi, O9,0. Gr ORT 95 0 GGle Ofer Wat, Cel, Gt. Joe ie Se°0 freon. OF OF ty leet Oc 1 UO vo ne 0G 7 Cone Io). ce sult menGy = Vost 6 b.S oeo-¢ Tees. 6656 foo. Toa is OTE soln, SVG HGrO9. -Gb.9. SGek. eOse> COG er ce gee [eg t6- “Vio. cor SGees Es Ges Wr G2 e008 Ge"0o- 66S 0G Ore are cote OS Ge TOPE go°te Io's GO dy-or- Seo Wie: ob lis 66°11 OO" ert OG5O) G@rc? Oe Gra —9S°¢ £9°0e. eG 36 eo te «Gt. ee Lh go°Gt @0re SOS. Goro. ° TerO cit -SG.O- ~SS"t je-Ge =e? -SOrnG EO ORE «SO Gh: — 60'S co'oL «6 6's COGS «=~ GFO) 9G) SOE: CEG Oe 1O°OT = Goss co-P Gent 1950) GeO — Taig 09° 1 G0"Ec GOS Ose Si.) Sh SE OG cE? OG"te -S2°= -Go>y- eG-0' Sono -@oc1 “Jig - 9S°T £6°6t- SIRO. -667e - “oT east dee Gere 19°C Goce -e6°0— C676. “70-0 ce 0. 360'9' 9651 dace [yi Sk we ole On ee) RaSh OG= —Ssiite 6/3 Ay O66. Ges. GCs) OO dere, =6Ort The. 9ore EGG -GOre Shee O10 en th =cO. 9 Oc S = or: ue SG LOT Ol ol 056 “Ole OS 4 = co 669. ror OG-ce ES 85°". 91 T- OT 96°96 99° TT. G52 AS €9°eT T6°S Meet TO’E Te1LOL ‘oq * AON “290 ‘9idec ‘Sny Atop oun "pyquoo “q eTqQeL « . M+ UNA O MA \O MMNUN AO FAN ANMHAMON AT as a ro co ° ° e ° ° StUAAAOAGEy LN LONREE ARE AMOS SNUANHASUNANANA Ot UAnUdHACOOOSO +StTMmWNOrRMNmOA OWN OVCD\O\O OO ONO OF DHDODODDGCOOCO . ° s . . Se+FNNOOHDAODOOHONATO MOO ao OFADMNDOANDDOHAMIDS COMAIO SF OVS INO) iy CO O00 8a ge" hb FINoUOK 66°9 eynuery oS'h, eTany 40'S BUeWa Qy Gc*6 BUueTeN Wea CMeIET, Ex°S sueTeqy neh FOyerey HUTT Fate. Tzeyng 99°2 UTHeEW 9T24TT CO6T seIOLYy eueuUey, 2070U0 neuNnsyT N nieg BoneytqedL TINOUON eynuetly eTIny eumeuls qy BUSTEN BNCIBY, Sueteqy Toyesren TIVLT.reyng UTYEN STUITT HE6T 1T°9 seIOIY Gee eueuey, qT OT BOYZOUD se : 3 SerOte ul *P,Juoo “CLOT ‘uer puertsyt . . . N+t++UNHNOHDOOOCOOOCSG MmOUWO ANMUNAQAHO a) ‘| aTqeL UT pue ‘azey peyyTWO useq avey sTeqO> ATaeek “saseo Yous UT °4T UO Byep OU ZO ‘TTejUTeI OU PSYSOTPUT YJUOM e TOT eLEP JO aouesqe TeyZeyA snoTAgO jOU sem 4]q ‘saseo JO AequMU e UT» bors ONY = Cu eG ele cece oO Occe “1c O de" qo" L =o ene rey, GQ°9T OTS = = Gore” Jeg Orr G90 SFG ~Gr-O THs Sec suereqy cou WOOL coro =Jo°O =Ge0O cee ie 2007s —Go°O €0-7 HET Gtt» O779 Toyereyy Woo et (On —GG70 (yO “dire “GO, SE “eo% “Ser Eos “GLY 00"S Tareq Treyng Seo gGir weal O96 GhO Gio Gall Feel Sy 2 EET fO"G OO°E “Si-O UENENSTa4aTI QE6T = Gls <6: -O7-f ZG62R “oR°e Ota -€O°r - ene T = 20 — 29021 — se-c Oo EL eT any NiGsee 6.0 ~GLiT “G20 80:0 Gle’s “66:6 “Th0 60°O eto lst Geri oO9:0r €9:n Gi‘er @A°TTt 6e"e ~ En Ht eT Any gets TOOL (070 40040 OTT 0950 Were -@0°0, WG:ec wets: ~ES:TT -L6°TT Gort euremaqy Roce sgo.g =Ors0l iT Oil <9956 seuee atolim T6°G- 06°6 -46°9 . -he-€ = EF OT euRTeW mise 206%) aGocct ce WoT “OGNGr WHS 26: Gorm: terme DS:9 99:6. <4LG°er enerey, TO*AG — Igo “Seat 46G9€ Gir 00°O = GieO> TiO —OO2G> "R86 eynuesy LS'h Shit 7, O- Oc-0- OO8l- Gees GENO SieO: O10" GIc0s- ETO 000°0 . -Sea"0 BT Any 69° Gese— 7620- Oe-0=— OF 16 02212 OSFOk Gr0: 00°00 ~G0"O ~ giz0s 500°0= 00°0 eureulaqy 66°6 oe Tese= Glee Sit. —Jsect -60°0 Hee one O050, 20205 O010— 62-06 BURTEY GE°ST pec Clee - 008 Cet le— O0ce = 6520 Te Oe tee Or— 2080. Cr-O. JO. c670 eMeIeY, Cyrel —Gses- OteTs GG2ie -CEtts 0080: — O521r - ci{Os 95:0 ae 6520. co.0 —_ba20 sueteqy 1e°Gl Gere: Hae le Hees = Gatos tee, Gee CIO. ET. «CEG OS TSEO OTE = ST Teyerey Garoc- @ent_Ce.c- cO-G-- coun Of SG 2 Ghee tire 109 Ofte Ge Ted TieqTregng GeeOGe tie CO 6 = cosh cee ere eso co, Orec O17 E6°E — Te) 0b 0 UREN STs OS6T feGye. GeO 66°0, O84 09-0 SG; =O) thOe O20 46°0 91:9 6°09 ~~ 2G°OT ae10Iy QT'O = GS Jeo tet Clie oe ced = HOEGss O28 S02 cr: eueUeL, Og-th., 9r-0:., oO: 16:0 Jeo: GO 18 OS Ore ost ford i Son: — Lesce 207 0U0 68’0 2 = OO" test Gerce GeO - Cente GO 2s oxS = SO GE neunyTN 6c Or TEO O1:0-.-Co-O- Of 0. Jer cot Gow Gee Geo J6o- Mes .olege nieg €2°0 S = tO hie @tee = Olt orc Fe°O 6L°e ee 69° SE Boney tqeL 4T‘*O < 42°0 O°T = Torce COm oO: OFee TOre 9°S 96°92 TqanouoHy GL’°O = EO -T9s0, (~GieOr Gee. 9S1 Bete ON 6T°ST eynuety GEG. HOO, OS Gre 060 TRG cha Gie Shne Lote % E Teiny 29g'0 = fet ) 920 “JSLOr . GEG. ceo = Lys met COnre Get ememeqy 99°0 = Gort 96r0- JO'O.. Ott -4Ore Git .¢e a Le Os TES Ts. BueTeN Cueta = Jeo Gee GeO — aiteg BMBIBT, GGreh uatOr9 abc 0 . 620 sorG boc SG0"G SO0TS sacrS eorS gmlesO areo segrt Ssuereqy lerGe SOLO ge70 —cG 1 re Moet Sole Goe Liat agi atiO si 0 Scale Toyerey Sento = 1GO oro <6 Tl oo°S “co = 19 fo 9 9 2 OL OL 2750 Oi bonne Trey Treyng Biro =r O00 Goo = OLS 962. ory —=GG'c —-GG°9 ~ TL"G Ort os°€ gf°T UTzEW Tae G6T Telco, e2eq StoN S00 Sades “ony Ayne -‘erne Aey rady yoreN cqaq ‘uUer puelt sl "p,yUo0D ‘qd eTqPL Table E Annual rainfall totals for the Gilbert Islands Island 1905 1906 1907 192425 1925-26 1926-27 1927 Little Makin 1.5. 179.04 Butaritari 178.19 172.10 Marakei 38.63 54.78 Abaiang nly ga 52.84 43,87 23.23 SHINS at Tarawa 23.13 63.80 Maiana pl rie 64.12 16.61. 3/00 Abemama 47.26 103.00 Kuria 24.20 10.80 Aranuka 23.15 Nonouti 5.76 Ba) si Tabiteuea 29. 39 S1.16 Beru 12.78 p87: Nikunau 42.72 61.39 Onotoa 14.63 13.68 Tamana eee 19.64 Arorae Bio ot) 17.90 Island 1927-28 1928 1928-29 1929 1929-30 1931-32 1932 Little Makin 117.33 96.33 Liens (4.93 60.70 Butaritari 125.02 Tbe rtsul LOTTO 105.0% 99.62 | Marakei 97.30 63-12 103.79 TSS MASHER SS A Abaiang 97.20 66.35 108.59 67-61 71.39 ) Tarawa 62.311 67.40 he 25 74.99 97.88 40.78 HH Maiana 76.85 31 07 93.04 26. 38 Abemama 79.85 28.48 78238 22.91 Kuria 69.65 30.88 82.32 38.71 Aranuka WOLes 32.82 77-76 41.50 Nonouti 46.53 17.92 65.83 Baee Tabiteuea 40.03 16.14 45.93 50.66 Beru 70.45 20.75 58.23 Nikunau 61. 4h 15.24 58.72 Onotoa 69.80 23.75 66.85 Tamana 65.30 26.97 63.37 Arorae 60.53 28 4h 67.50 Sources: Compiled from 1, 2, 3, 4, 7, 8 Island Little Makin Butaritari Marakei Abaiang Tarawa Maiana Abemama Kuria Aranuka Nonouti Tabiteuea Beru Nikunau Onotoa Tamana Arorae ~ Island Little Makin Butaritari Marakei Abaiang Tarawa Maiana Abemama Kuria Aranuka Nonouti Tabiteuea Beru Nikunau Onotoa Tamana Arorae 1932-33 82.39 115.45 62.35 39.69 50.25 28.25 25.94 32.69 35.83 23.69 29.94 20.88 14.83 31.50 bh 58 32.25 1940 110.68 LIBS Y 151.62 139.97 128. 36 128.02 128.32 108.05 111.65 115.10 128.92 140.42 148.16 155.45 129.38 128.14 1934 82.86 107.98 57-63 49.98 33.47 22.55 19.93 23.56 23.08 18.67 16.03 25.27 15.66 20.63 17.99 23.08 1944 37-95 Table E, Cont'd. 1939 90,28 95.67 69.54 59.01 4.88 49. 34 37 . 36 30.02 42.35 31.45 37. 4d 38.89 26.24. 45.86 50.21 51.18 1945 27.39 1936 88.90 154.19 87.05 97.01 78.43 13.99 54.32 59.48 53-76 58.56 WO.168 61.52* 72.01* 64.57* 46.10* 53.59% 1946 50.19 157-97 81.10 97-48 97.76 115.19 95.8h 15.09 80.23 83.29 *Figures taken from 3, those in 7 are discrepant. 1937 67.74 60.99 41.90 35.05 33-13 22.51 26.02 25.78 oh 46 31.56 38.51 197 50.54 156.29 63.96 Te 49.96 30.51 1938 108. 33 52.74 hh 32 9-15 14.61 17-09 1948 73.47 145.03 110. 35 115.49 114.26 62.12 66.89 17.26 87.49 69.63 99.50 76.63 81.95 Island 1ghg Little Makin Butaritari Marakei Abaiang Tarawa Maiana Abemama Kuria Aranuka Nonouti Tabiteuea. Beru 48.29 Nikunau Onotoa 41.80 43.34 119.80 M27. 20 91.18 Tamana _ Arorae 1950 50.29 56.85 19.81 12.43 15.35 Table E, cont'd. 19D 1952 153.35 131.42 153.83 149.07 103.85 88.88 88. 33 97.63 103.88 13-55 101.21 79.76 78.90 50.15 74.72 22-92 73.26 49.06 61.45 40.90 56.21 40.13 61.35 45 43 50. 33 41.04 74.68 36.90 74.30 60.09 82.82 67.93 1955 49.78 60. 39 25.67 42.99 25.42 17.69 12.91 8.84 TnL .Gis 10.94 17.22 15.24 16.75 17.07 30.77 22.55 Averages 100.23 121.50 (Ce (3.p0 64.02 Dasa 53.04 48.37 49.72 4.3.18 40.98 45.29 INI AG) 45.83 50. 39 52.30 ' ” Nos. 61, 62, 63, 64, 65, 66, 67, 68, 69, 70 ATOLL RESEARCH May 15, 1959 BULLETIN 61. Long-term effects of ' vadioactive fallout on plants? by F. R. Fosberg 62. Health and sanitation survey of Arno Atoll by John D. Milhurn 63. Report on a visit to the Chesterfield Islands, September, 1957 by F. Cohic 64. Canton Island, South Pacific (Resurvey of 1958) by Otto Degener and Isa Degener 65. Some marine algae from Canton Atoll by E. Yale Dawson 66. Notes on the geography and natural history of Wake Island by E. H. Bryan, Jr. 67. Vegetation and flora of Wake Island by F. R. Fosberg 68. Additional records of phanerogams from the northern Marshall Islands by F. R. Fosberg 69. Contribution to a German reef-terminology by Georg Scheer 70. Atoll news and comment Issued by THE PACIFIC SCIENCE BOARD National Academy of Sciences—National Research Council Washington, D. C., U.S.A. ATOLL RESEARCH BULLETIN 61. Long-term cffects of radioactive fallout on plants? by F. R. Fosberg 62. Health and sanitation survey of Arno Atoll by John D. Milhurn 63. Report on a visit to the Chesterfield Islands, September, 1957 by F. Cohic 64. Canton Island, South Pacific (Resurvey of 1953) by Otto Degener and Isa Degener 65. Some marine algae from Canton Atoll by E. Yale Dawson 66. Notes on the geography and natural history of Wake Island ti by E. H. Bryan, Jr. 67. Vegetation and flora of Wake Island | by F. R. Fosberg 68. Additional records of phanerogams from the northern bi Marshall Islands : ‘ih i by F. R. Fosberg 69. Contribution to a German reef-terminology iy by Georg Scheer na 70. Atoll news and comment Issued by THE PACIFIC SCIENCE BOARD National Academy of Sciences--National Research Council Washington, D. C. May 15, 1959 ACKNOWLEDGMENT It is a pleasure to commend the far-sighted policy of the Office of Naval Research, with its emphasis on basic research, as a result of which a grant has made possible the continuation of the Coral Atoll Program of the Pacific Science Board. It is of interest to note, historically, that much of the funda- mental information on atolls of the Pacific was gathered by the U. S. Navy's South Pacific Exploring Expedition, over one hundred years ago, under the command of Captain Charles Wilkes. The continuing nature of such scientific interest by the Navy is shown by the support for the Pacific Science Board's research programs, CIMA, SIM, and ICCP, during the past eleven years. The Coral Atoll Program is a part of SIM. The preparation and issuance of this Bulletin is assisted by funds from Contract N7onr-2300(12). The sole responsibility for all statements made by authors of papers in the Atoll Research Bulletin rests with them, and do not necessarily represent the views of the Pacific Science Board or of the editors of the Bulletin. Editorial Staff F. R. Fosberg, editor M.-H. Sachet, assistant editor Correspondence concerning the Atoll Research Bulletin should be addressed to the above ; c/o Pacific Science Board National Research Council 2101 Constitution Avenue, N. W. Washington 25, D. C., U. S. A. ATOLL RESEARCH BULLETIN No. 61 Long-term Effects of Radioactive Fallout on Plants? by F. R. Fosberg Issued by THE PACIFIC SCIENCE BOARD National Academy of Sciences--National Research Council Washington, D. C. May 15, 1959 Long-term Effects of Radioactive Fallout on Plants?2/ by F. R. Fosberg Botanist, U. S. Geological Survey A number of fallout surveys have been carried out since the widely publicized Castle Bravo bomb test on Bikini Atoll on March 1, 1954. Three of these were under the auspices of the U. S. Naval Radiological Defense Laboratory (NRDL). Among the results of the first of these sur- veys was the establishment of a series of stations in the Northern Marshall Islands in which the amount of fallout varied, with radiation doses ranging from almost none to doses dangerous to human and animal life. A series of stations selected for subsequent resurvey are listed in order of increasing dose received expressed as total dose to infinity. This figure was obtained from the intensities in roentgens per hour on the day after the shot was fired. These rates were calculated from the earliest actual readings available and are given by Dunning (1957) on a map on page 2. The formula 5 x 2h x 4/hr at D 41, where D #1 is as- sumed to be D a 24 hours, is used to calculate total dose to infinity. The time of arrival at Rongelap Islet was about 7 hours after the shot. The time to the other stations can safely be assumed to have been a Bye the same or longer, which would give a maximum error of 22 percent. List of stations: ikep ALOLE Likiep istetws 7. 2". se. +s ». scarcely any Unit aco. Ubi is VEG es tees ane oes Pilineinde Atoll, Sifo Isleg es... ee i255 Rongelap Atoll, Rongelap Islet ........ oe 336.r Rongerik Atoll, Eniwetak Islet .......... 365.r Rongelap Atoll, Eniaetok Islet ..... es aes 816.r M Iu, Kabeides Telet mane ii Wiis 182h.r M MW eiGegen Tsdle tt "0a" Be ites ee, SROONE These figures, of course, say nothing of the proportions of partic- ular isotopes present, except that since 7 hours elapsed before this fallout was deposited, the longer-lived ones will have been relatively abundant, which is a very important fact, physiologically, as it is these which will be responsible for any cumulative effects. During the first two NRDL surveys no visible effects on plants were recorded, though some of the food plants tested showed considerable accumulation of radioactivity in their tissues. The author of this note accompanied the resurvey of these islands in February, 1956, in the capacity of coral atoll ecologist, but most of his time was required in helping with the collecting of samples for uf Publication authorized by the Director, U. S. Geological Survey. 2/ Information supplied by Mr. Sam Rainey. BMITHSONIAN MAY 15 1980 INSTITUTION il age analysis. Because no gross effects on plants were observed on the ear- lier surveys and since radiation damage to plants was not known at the radiation intensities existing on these islands no provision was made in the planning of the work for time to observe such phenomena. Time was not available for careful study even after it became evident that the vegetation on some of the islets visited was not normal. Casual observations were recorded, mostly of a general ecological and descriptive nature, in the course of sample collecting. However, until members of the party recognized that the poor condition of the vegetation on Gegen I. had not been evident the previous year, no serious attention was paid to these abnormal phenomena. By this time it was too late to make any systematic study. Subsequent study of notes and photographs made on the 1956 survey has indicated that suspicion of cumulative effects of radicactivity is justified and that further field study should be directed specifically at this problem. Until specific studies can be made, it seems best to publish the available data, even though they are inadequate either to establish or to disprove any cumulative radioactive effects. Perhaps if attention is thus called to the problem, someone in position to make further observations may be induced to collect data that will settle the matter before so much time has elapsed that all effects are masked. In a report to the Atomic Energy Commission in 1950 St. John and Biddulph described visible abnormalities of various sorts observed in August 1949 in plants growing within a mile of the site of the 1948 atomic explosions on Eniwetok Atoll. Some of these were interpreted as persistent physiological effects, others as "phenotypic mutants". The plants affected were the following: Fleurya ruderalis (Forst. f.) Gaud. Ipomoea pes-caprae (L.) Roth Ipomoea tuba (Schilecht.) Don. Morinda citrifolia L. Portulaca oleracea L. Scaevola sericea Vahl (as S. frutescens (Mill.) Kr.) The environments studied on the 1956 survey were very similar to those in which the plants studied by St. John and Buddulph grew, but the intensities of radioactivity were many times less. On the 1956 survey, attention was first attracted by Suriana maritima L., a small-leafed bright green shrub, which showed dead branches or completely dead bushes wherever observed (condition not re- corded on Rongelap Islet, as attention had not been drawn to it at the time Rongelap was visited). - Until Eniwetak Islet, on Rongerik Atoll, was visited only casual attention was paid to this phenomenon. Here, however, the Cordia forest -3- which covers a large part of the islet appeared a dark gray color from the air. This was found to be due to a dying back of the twigs over the entire tops of the Cordia trees for as much as 30 cm. or even more and a complete loss of leaves from the upper parts of the trees. The Pisonia grandis associated with the Cordia was defoliated but flowering, a condition which could be normal in a severe dry season. On Gegen Islet, Rongelap Atoll, subjected to the most severe fall- out, almost the entire vegetation was observed to be in very poor con- dition. Cordia and Guettarda, two of the most important shrubs, were dead or apparently dying. Pisonia trees were entirely defoliated except for mistletoe-like clumps of abnormally dark green leaves of an appear- ance not previously seen in this common species. Lepturus repens, the commonest herb, was completely dead except for a few tiny seedling tufts on the beach ridges. All Fleurya ruderalis plants observed were almost dead. Some coconut trees were dead. The other species present, with the exception of Scaevola sericea and Tournefortia argentea, were in un- healthy condition. The latter two species were in generally good shape wherever observed. The following two lists present the complete observations, so ar- ranged as to bring out any correlation with fallout intensity. 1. Plant species in which abnormal appearances were observed, occurrences by islets in order of increasing fallout in- . it tensity. | Suriana maritima L. Utirik I. - abundant, very few dead or partly dead plants seen. Rongelap I. - no specific observations recorded. Sifo I. - mostly healthy but some bushes dead. Eniaetok I. - mostly dead. Kabelle I. - many dead bushes, no live ones seen. Gegen I. - none seen. (Similar partially dead Suriana was observed in 1956 on Enyu I., Bikini Atoll by Mr. Sam Rainey (personal communication). ) Pisonia grandis R. Br. Sifo I. - somewhat defoliated but not strikingly so. Kabelle I. - almost leafless. Eniwetak I. - leafless and mostly in flower. Gegen I. - nearly to entirely leafless and flowering, but with compact clumps of abnormally dark green leaves that somewhat resemble mistle- toe from a distance. Cordia subcordata Lam. Eniwetak I. - appeared dark gray from above, tops bare of leaves, upper twigs dead to 30 cm. or ah more from tips, trees generally rather de- foliated, more so in west than east half of islet; a few of the bare terminal branches putting out weak sprouts. Gegen I. - mostly dead or almost so. On other islets, normal so far as observed. Cocos nucifera L. Uterik I. - normal.. Rongelap I. - normal. Eniaetok I. - normal. Kabelle I. - very yellow. Eniwetak I. - no record of condition. Gegen I. - some trees dead. Guettarda speciosa Tis Gegen I. - dead at tips in open scrub; in forest largely dead or severely dying back from tips. On other islands, apparently normal so far as observed. Fleurya ruderalis (Forst. f.) Gaud. Gegen I. - almost dead. On other islands, apparently normal so far as observed. Lepturus repens (Forst. f.) R. Br. Gegen I. - common but completely dead except for a few tiny seedling tufts on beach ridges. Elsewhere apparently normal so far as observed. Boerhavia diffusa L., B. tetrandra Forst. f., Achyranthes canescens R. Br., Pandanus tectorius Park., Portulaca lutea Sol., Cassytha filiformis L., Ipomoea tuba (Schlecht. ) Don, Tacca leontopetaloides (L.) O. Kize., and Terminalia samoensis Rech. all in poor condition on Gegen I. but nature of damage not recorded. All apparently normal elsewhere so far as observed. Islets visited, in'decreasing order of fallout intensity, with plants in which abnormal appearances were observed. Gegen I. Achyranthes canescens Boerhavia diffusa Boerhavia tetrandra Cocos nucifera Cordia subcordata Fleurya ruderalis Guettarda speciosa Lepturus repens Pandanus tectorius Pisonia grandis Portulaca lutea Suriana maritima Terminalia samoensis Kabelle I. Suriana maritima Cocos nucifera Pisonia grandis Eniaetok I. Suriana maritima Eniwetak I. Cordia subcordata Pisonia grandis Sifo I. Pisonia grandis (only slightly defoliated, probably normal) Suriana maritima i Rongelap I. No species noticed in abnormal condition, but this was first islet visited and perhaps damage was unnoticed as attention had not been directed toward it. Utirik Islet Suriana maritima (not strongly damaged) These data are inadequate and are therefore only suggestive. It is by no means claimed that there is any causal connection between fall- out and the phenomena observed. However, the clear observation that the poor condition of the vegetation was more intense on the islets where the fallout was more concentrated suggests a relationship. This seems to be of sufficient interest to warrant a careful study. Such a study should determine, first, whether these effects are still observ- able, and, if so, what the actual nature of the damage is, whether there are significant concentrations of radioactive substances in the tissues, whether there are any notable abnormalities in recovered plants, and whether any young seedlings show abnormalities. For quick reference all records of plants from the three atolls most seriously affected by the 1954 fallout (except Bikini itself), Rongelap, Rongerik, and Ailinginae, as well as all available records of Marshall Island plants affected or putatively affected by fallout =oe or other radioactivity have been assembled in the following table. Records indicated by T and by Fe are substantiated by specimens de- posited in the U. S. National Herbarium. Many additional species were seen on the NRDL survey but not collected because of lack of time and on the assumption that the species would have been picked up by Taylor at least on the same atoll. This did not always turn out to be the case, but the Taylor and Fosberg collections complement each other fairly well. Important extensions of distribution are reported in a separate paper (see Atoll Res, Bull. no. ie with citation of speci- mens. A Legend of symbols used in the table: Habit of plant: -- herb creeper or vine -- shrub -- tree tag Bp t ‘ Abnormalities or reported effects of radioactivity: Sp -- St. John, 1950, report of persisting (physiological? ) effects. Sm -- St. John, 1950, report or ecation. Bp -- Biddulph, 1950, report of physiological effects. * -- abnormality recorded by Fosberg, 1956. U -- abnormality recorded by Fosberg from Utirik, 1956. R -- abnormality observed by Sam Rainey on Bikini, 1956. Miscellaneous data on flora and collections: T -- Occurrence reported by Taylor, 1950. F -- Occurrence recorded by Fosberg, 1956. c¢ -- Herbarium collection by Fosberg, 1956. n -- Material collected by Fosberg for radioanalysis at NRDL, 1956. € -- Material collected by Evan C. Evans, III, for radio- analysis at NRDL, 1956. : . ot & | eee : a mNoltVerse WNUTID | 2 ! Loe aa | | a | | en) of | wea | wot | adh) | 9 EAE : | oe : | | : Ma | a re eS ee Hite esouko STTAysTAqUTY | | i = i i i i i oud oug | ouq ie oe iu: S4NTOAUT eerenuy | | oe { i i $ t i } : ! | : ay _ ugdg a eyeutTyos snayouep : t { : : , ! : : Lj ae eBOTPUT SUTSNeETA : ee Ste | ips eed [ou Totti a 1p ead ek ck | sTTeuoTayueqdes | ' } i leat “IeA suoeder snangdey | | ' { nn t nee sci 8 ee : us euyoeqoroTM * eA | ! | gsuetanid eTreLtstq ' i i ie] [Eel ag ol | cret | au! st[iqeme stisoiesy H } | | ie ot i | i j i i re | | dg iy | BYBTJUT STzoTYO | | : ae Dag | i 08g jomd Lb) od | oF | 408g | i al snt103084 snuepueg | Uae od | | ! Ui ezapuedotoos umppodctog ea ee ead erp a rc eg vet ee ater ae eee ed Se memrune ie eS eee ee Bee a aie s 5 ee ee ee eee = i i ¢ 18 bp. ec | Q Se | fo} | @ Be le | 208 i o i o i © HO |e! | pte es } @ | La} i © H 1) kh =] © , ct aa | Pe es a i eS 4G | | See 3 eas eee iF, i H | : { Hecate i Onk | Hf , oH [ oH eas i rect 00 | | | : i= i ! Hh bs ig = CLUS | | ae | | | qe | coy | Trey [107 detesucy quetd go swell SeUuTSUTTIV | YTLIesuoy | uBe ay oug xOUY oud od ‘I OTS ~ TTOW SeUTAZUTTIV | | ak | He ali i OW G OUT os | of L | 5 bo} We sea oe te ea ae ee ae ® Ie 8 | Oo | ct + © a ce Tow YT IIBZuo}y i SS SS ' he nae ne ee meme ts T3UI0! ey a ee = ee } 9 me eaarpererennd meBiyn Fm rom j ‘I depesuoy 0) aS) 4 ) Fy t | | | [eo | | | jour i wa | | Ul LOX i f | oud i { | eee | | | : (= oi | | : | | | oat oi ce ye met ee ee i fc | gt | : fe) He (40) [eee eed ; et | 4B Ee i } | TTow “depasucg | jal edieoo1otTm BT Teaceue) | ie} ‘ds etutdteseeg lee iS % STWIOSTTTZ euzAssep rere tg STsusomes BoeTny10g i Poe | Igugds iy easodereTO eoetTny10g ae eoyntT eoeTnj,10g rio stpueis eTUOSsTd i | Hi - ’ xe iA erpuel7.e} eTaeyse0g xUS LA. eSnjitp eltAaecy.1s0g t . ue sueosoueo say uerkyoy { xuS YU stTeiepnaz efameta i . ae STTTAITS sndzres0 zy U SaptoteyedoyuoeT BddRI, 4 = a =a Reece nee Eee SE es Een oe ea cee ee ae Pes ee ara m6 me) hb t MO eS Lewy tar = HK Ele) go ¢ Se ci i ean ick Peoe ce o "28 ee ee ee ee eee quetq jo amex, ih | \ | | | OIG | | aseideo-sed esomodt ! i } a ia ro yak L | 4 | eTTOITYTsoddo etso7yo0 | | YI | L | Ts) L gO | 89 | stsusomes ST TeUTULZAY, } i { ! : : | be : Hpeee i | ! Ea | eresnfuos ezetnen7zg | ‘ ! H i ' t : es | : 84 | eTnproe stydmeg : i t ) 4 | | oat |) | | A | eMTXEM e4YTQamnong i ; i | f 4 { | : oY: a | a ekeded eotreg i { i : j i ‘ sie : | 4) umttéydout unt ytAydotep : 1 : i | | fe ed, Je poh) by a ese) XeTTRI PTS ! | (24 OW | | ca ant Te a | pd susqmooid e44azuMT.2y, i ! | ‘ i ‘ i i i oa aie | | i stsuzouty sntftydotty { { : i i H ' ! { 1 { H i i | 1 Od | ! [aa ereue vauepnog | | | | | | i #0 wb | pat bo | xf | | eT S| eUTITreM eueTIMs Sere ee eS 7 ak ie es ee eee ee = ee eee : t (2) } i tf i | ' . } ns poe 08 | ot eae . os o> & i 8G B. ee) t =a jo aE fee 2 ee) \ [a i lox i 0 Bist o7t o lane / & ; © 1 9 ee @ ; © Pao uee (2) Stays KH i { @ | i oa I ; tS ate : eee | ie poet ees eee es | & as) ae | a (ees sey | } oH lo b } =. es paces i kb ck |) OQ | ce i | ae a | | i Gace! = | | feck = | ee eae foes Sea eee | Seem ee eee een Sc nee Eee rt eR, pil pe React quetTd jo owe, TOW | TtOay Tlow detasucy SeUTSUTTIV ' ¥YTresuoy Bio oF aT aro oud TTO1 QeCUTAUTTTV | L ad ee Vey 4 Ug | | | | | : | } “uj 38 | 'yod DL ea: (oom eee | oO - 4 = ee se © | [oF | See Es | | |. prea YTzo99uo0y | aa ee | | di oh i eee ee } | od ee | | | ; OF 1 Wb |; ww { ! ue iNet nh | 3 [ne 7 oa | bo H i t ' |G | Of | ao | fa | : (eerie A | te reek oe Se er ee oe eae (@) 2) & ra i | tof | al i ee ; @ i H it) i @ (echt per. ha etd are | Bb 6. © | bearer eet | | | ieee | aR eee Nas Sane cates | Tleay detadu0y or | xUt | ye xd | x1 A wg wg ug is! es eucar xdqus 6A - SOTVTTeusouaqe: YANOIS Jo 4Tquq. BIOTITC SLTTSPEM eooTsies eToAseog ST TOST4AZTO epurszop esotosds epireyjoeny TIszeyyns -1e0 uMustjUuBIEepnesg SUISUT UMAIPUSPOISETO eoqussIe eTLIOTOUmMO], eyepazooqns etpiz09 eqny Beaowody quetTg Jo sure] Satake REFERENCES CITED Biddulph, 0., 1950, pp. 55-93, in, Applied Fisheries Laboratory, University of Washington, Report No. UWFL 23, (AECD3446) (unpublished). Dunning, G. M., 1957, Radioactive contamination of certain areas in the Pacific Ocean. 1-53, U. S. Atomic Energy Comm., Washington, Die Gr Fosberg, F. R., 1959, Additional records of phanerogams from the Northern Marshall Islands. Atoll Research Bull. 68; 1-9, 1959. St. John, H., 1950, pp. 37-54, in, Applied Fisheries Laboratory, University of Washington, Report No. UWFL 23, (AECD3446) (unpublished). Taylor, W. R., 1950 Plants of Bikini and other Northern Marshall Islands. 1-227, Ann Arbor. ATOLL RESEARCH BULLETIN No. 62 Health and sanitation survey of Arno Atoll by John D, Milhurn Issued by THE PACIFIC SCIENCE BOARD National Academy of Sciences--National Research Council Washington, D. C. May 15, 1959 Health’ and’ Sanitation’ Survey! of Ayno Atoll, 1950)/ by john By Midwawne! sce In the late 1940's, the United States Navy assigned the U.S.S. WHIDBEY (AG-141) to conduct health and sanitation surveys of the Trust Territory Islands. The WHIDBEY arrived at Arno Island, Arno Atoll, on April 30, 1950 for this survey. The following summary has been abstracted from their report: Arno Atoll has the largest land area of any atoll in the Ratak chain of the Marshall Islands. It is composed of 25 to 30 low islands and coral reefs which seem to be the remains of a crater rim. These form a circular barrier protecting a large lagoon. Many of the islands of the Atoll are densely covered with coconut palms and pandanus trees. A coral reef which is difficult and at times hazardous to navigation pro- tects the entrance to the lagoon. | Two major villages exert their separate spheres of influence. Arno village on Arno Atoll influences the people of the surrounding smaller islands. Ine village, on Ine Island, a long narrow coral island in juxta- position to Arno, is more important, being the seat of the Island Trading Company for the Atoll. Natives living on smaller islands of the Atoll while making copra are listed for census purposes as belonging to either Arno or Ine village. On festive occasions such as Christmas and the Fourth of July, which the natives have adopted, the entire population of the Atoll gathers on Ine for feasting, celebration, and even baseball at the Ine ballpark. According to the Civil Administration Government at Majuro, the 1946 census of Arno Island was 256 and of Ine Island 302. The U.S.S. WHIDBEY located only 194 persons on Arno and 206 on Ine, including three visitors from Kaven (Maloelap Atoll) who had been examined on the ship in 1948. A determined effort was made to examine every native. One of the medical officers of the WHIDBEY visited each house or hut to induce the occupants to visit the other medical officer at the town house in the municipal square. It was the opinion of these medical officers that every native in both villages and their environs was processed and ex- amined. Time and the difficulty of locating those natives on the sparsely settled, small uncharted islands made efforts to examine the remaining population impracticable. 1. Abstracted from original report by members of the Preventive Medicine Division, Bureau of Medicine and Surgery, Department of the Navy, Washington 25, D. C. 2. Member, U.S.S. WHIDBEY (AG-141) Survey Team. 3. Opinions expressed herein are those of the authors and are not to be construed as official or as representing the opinions of the U. S. Navy or naval service at large. ec It was noted that the Marshallese are a very itinerant people who have little hesitation in undertaking sea voyages in an outrigger canoe. One of the natives not examined was a Chief Magistrate said to be visit- ing Majuro. This suggested to the WHIDBEY group the importance of planning surveys for simultaneous coverage of whole island groups rather than individual islands. Despite the failure to locate and examine a portion of the Atoll population, the WHIDBEY group were of the opinion that a representative sample had been examined. "The Marshallese on the Arno Atoll welcomed the WHIDBEY with curi- osity, delight, grace and hospitality. They showered us with their home- made leis, and gave freely of their limes, papayas, bananas, coconuts and pandanus. At no time were any of the beach party thirsty, being lavishly supplied with coconut milk. For the people, especially the young and the ancient, the ride in our dukw and the visit to the ship were holiday treats. School was even called off (and this probably occurs at the slightest pretext). The authorities there freely cooperated and showed our sanitation officer everything, hiding nothing from view. Even the dogs were startled at the sight of the monstrous dukw rolling along the primitive, Japanese constructed island roads. "The young men demonstrated their native game of jog jog (phonetic spelling). They played this game, which is reserved for the men only, with great frivolity and mirth. The game is similar to our red rover and chug-a-lug and is played with a square ball made from pandanus fibres. The men stand in a circle and the man out stands in center. One man be- gins by throwing the ball in the air and kicking it with the sole of his everted and flexed right or left leg. As the ball approaches another man, regardless of his position in the circle, he must kick it similarly and so on the game progresses until one man misses when they start over. At the time the ball is first kicked a single clap of the hands in uni- son by all starts. After the third kick the clapping is coupled. Evi- dently the center man merely has the duty of keeping the ball in the air should it come to him for they never assumed or changed places in the circle. The problem of scoring is a mystery. "Arno Island is about 2 1/2 miles long and 1/2 mile wide possess- ing in addition to its coconut and pandanus trees, a thick underbrush of vines. There are about 3-5 acres of arable land but cultivation of cleared areas would be difficult. Ine is a long narrow island (13 miles long and 1/4 mile wide) with a topography similar to that of Arno Island. "The government of each village is democratic in form. The Chief Magistrate is elected for an indefinite period. In the Marshall Islands the Chief Magistrate's influence over his people is directly propor- tional to the extent of their Americanization, being less where American influence is greatest. There are also a registrar and a scribe of Arno Atoll, who reside on Ine. Whether or not each native understood the humanitarian instincts propelling this survey and just what he expected as a result of the survey is unknown. They submitted to processing and examination without question. "The native economy of Arno as elsewhere in the Marshalls depends directly on the sale of copra and handicraft. In dollars and cents, ~3- there are no apparent millionaires in Arno, but in spirit and health there are many, and perhaps proportionately speaking, many more than in the U. S. A. Their diet consists of home grown coconuts, breadfruit, pandanus, bananas, papayas, fish and pork and is varied slightly with imported rice, sugar, flour and canned corned beef, milk and beef. "Phere is one school on Arno Island with 30 students (6 - 14 years old) and a missionary school and an elementary school on Ine with 47 pupils (7 - 20 years old) pursuing general studies with some sanitation principles being added to the curriculum. All natives speak and write Marshallese but only 8 adults understand English. All the natives were Protestant, holding Sunday services with a native minister from Ine. The Congregational Church, U. S. A. has established a mission school at Kusaie in E. Carolines where the native ministers (usually the Magistrates) are trained. The Bible has also been translated into Marshallese along with many Protestant Hymns. Not infrequently, when you offer a cigarette to the Magistrate, the interpreter will say, "He no smoke. He Missionary!' "Medical facilities on Arno Atoll are searce. A new thatched dis- pensary is being built in Arno village. Obstetrical cases are delivered at home with little regard to asepsis by a midwife or the native practi- tioner (who although trained at Majuro is little more than a medical aid). The parents and immediate families care for the young, the ancient and insane. Tuberculous and leprous cases are left free to roam the Atoll, when found. There is no understanding of the principles of contagion and no facilities for isolation. The treatment of yaws (oftentimes being misdiagnosed as yaws and actually impetigo or pyoderma), of skin diseases and minor surgery conditions is left to the practitioner. Most of the people have been vaccinated for smallpox. Other immunization procedures are not recorded. No epidemics or water-borne or milk~borne diseases have been mentioned. "The water supply is derived from rain water and although adequate is considered contaminated. One 20-ton concrete cistern built by Germans was found on Ine. Half of the cisterns are uncovered and contain trash. All the wells are uncovered and do not exclude surface drainage. Some of the water is boiled. Each village either defecates and urinates in the available unscreened latrines or on the beaches or in the interior. Living areas are dirty and unpoliced. Garbage is disposed of, however, in sanitary fills or by burning. Housing is of thatch and although not too clean, is adequate. "Rodents and flies are numerous. Some culicine mosquitoes were noted on Ine. There are no control measures for rodents. Flytraps and sanitary garbage fills are fly control measures. There are no known plans to cope with disasters or epidemics. DDT is in use to control mosquitoes on Ine. "Vital statistics (births and deaths) are filed in certificate form by the practitioner at Majuro. The scribe keeps record on Ine of mar- riages, divorces; court records, births and deaths for Arno villages. "For the year 1949, the death rate at both Arno and Ine was 4s / thousand, considerably higher than the all-time high in U. S. A. of 23.0 AS in the last two decades. The leading cause of death at Arno is reported as tuberculosis but this statement may be challenged due to the poor vital statistical record and lack of knowledge of different diagnoses by practitioner. "The native population of Arno definitely surprised the survey team for they presented a summary picture of excellent health. While their health facilities are relatively primitive, environment and time seem to be wedded in a union to protect these people from the vicissitudes of our modern civilization. The lack of time to complete thoroughly the pre- ponderantly detailed survey examination of these people (3 days were al- loted per village) and the language difficulties do not insure accurate statistics. The figures in this report represent, however, the best ef- forts of the survey to formulate an accurate picture of the health and sanitation of this island.” Some of the results of the Health Survey follow: Coccidioidomycosis and Histoplasmosis Approximately 96 per cent of the natives were skin-tested with these two antigens, children under 1 year of age being excluded. The test con- sisted of intracutaneous injection of 0.1 cc of a 1:500 dilution of Histoplasmin and 1:100 Coccidioidin. Results were read at 48 hours and no positive reactors were found. It was noted that the antigens had been in refrigerator storage on the ship for five months prior to use and might have deteriorated. Judgment of the team, however, was that these two mycoses were either absent or rarely occurred since chest roentgeno- grams did not reveal abnormalities suggestive of old infections. Tuberculosis Thirty-five per cent of the natives examined had positive skin re- actions to 0.0001 mgm of PPD (Purified Protein Derivative) intracutane-- ously. One~half of positive reactions. were in persons age 40 or over. All natives were screened with 35 mm. photofluorographs. Those with suspicious findings were further evaluated with 14 x 17 inch roentgeno- grams. Three cases of active tuberculosis, two men and one woman, were found. All were over age 50. Three additional cases were strongly sus- picious, one a child less than 10 years of age, and two women over age 0. All six had 24 to 4/ dermal reactions to PPD. It thus appeared that about 1.5 per cent of natives had proven or suspicious tuberculosis. No case of extrapulmonary tuberculosis was found. The survey group commented on the apparent lesser incidence of tuberculosis at this Atoll than in other parts of the Trust Territory and the discrepancy between their findings and the official report of deaths due to tuberculosis for the past calendar year. Yaws Serological tests were completed on 149 residents of Arno (76.8 per cent) and 143 of Ine Island (70 per cent). Positive results were obtained in 103 and 94 sera (or 66.2 per cent and 69 per cent), respec- tively. Twenty (20) cases considered to represent primary yaws on the basis of a peculiar raspberry-like lesion of the skin were found. Scrapings from these lesions were negative on darkfield examination. Six of the 20 had positive Kahn tests. These lesions appeared to be a latent phase of the primary stage for they were crusted, and the positive serologies explainable because of the lateness. Seventy-nine (79) others were considered to have secondary yaws on the basis of peculiar annular, mottled, prickeled, indented scars of the skin. Such scars had to be differentiated from those due to burns re- sulting from the native practice of night fishing by torch light. Seventy- four (74) of the 79 had positive Kahns, the remaining 5 being too young for an adequate blood sample. The remaining 116 natives with positive serologies showed no evi- dence of primary, secondary, or tertiary yaws. Neither was there any history or other evidence of syphilis, infectious mononucleosis, influ- enza, yellow fever, or malaria. These cases were, therefore, considered to have inactive yaws with a negative history. Two women over age 50 had bilateral swelling about the bridge of the nose and gave questionable histories of yaws at an early age. Both had negative serologies. These were suggestive of N'Gondou, supposedly a tertiary manifestation of yaws. Four men and women over age 50 had positive serology and a dilated tortuous aorta. All had either an accentuated aortic second sound or a low blowing systolic murmur at the base of the heart. None had diastolic murmurs or wide pulse pressures. None had a history suggestive of syphi- lis. These may have represented arteriosclerotic heart disease or syphilitic aortitis. Because of the endemicity of yaws and infrequency of syphilis, however, the Survey Team raised the possible connection with tertiary yaws since similar findings had been made on other islands. One case with bone involvement considered to be tertiary yaws was found. No cases with juxta-articular nodes or gangosa were encountered. The overall incidence of active and inactive yaws among natives was estimated to be 53 per cent. Syphilis The Germans and Japanese had reported the prevalence of syphilis in the Marshall Islands. The survey found four cases of clinically suspicious primary lesions, all of which had negative darkfields. One other case was thought to have congenital syphilis on the basis of history. oo The diagnostic difficulties between yaws and syphilis were recog- nized. The possibility existed that some cases considered to have yaws might be syphilis. Further diagnostic difficulties in regard to active lesions were posed by skin lesions of pyogenic or other origin. Leprosy No cases were found despite a careful examination. A number of na- tives were found with lesions on the face or trunk which might have been suspicious but lacked any sensory nerve changes. Acid-fast stains on scrapings from a number of these lesions did not reveal M. leprae. Five elderly people had ulnar nerve thickening but no sensory changes. All gave a history of phalangeal fractures or traumatic laceration of the palms. " Parasitic Infestation Stool examinations were made on 253 natives. Since these were not fresh, accurate data on the prevalence of Endamoeba histolytica or E. coli could not be obtained. Sugar flotation and direct smear examinations were made in each case. Intestinal parasites were identified in 53 or 18.2 per cent of specimens, these consisting of hookworm, Trichuris trichiura, and Giardia lamblia. Infestation in specimens from people over age 15 was twice that of specimens from people under this age. Hookworm was found in 7 per cent of specimens from Arno and two per cent from Ine. The re- spective percentages for T. trichiura was 5 and 9 and for Giardia lamblia 1.2 and 7.2. Ascaris lumbricoides was not found. Mycosis Approximately 26 per cent had clinical evidences of mycotic infec- tions, mostly, Tinea versicolor. The incidence was five times greater in adults than in children under age 15 and ten times greater in males than in females. Two cases of Tinea imbricata were seen on Ine. Other Approximately 4 per cent appeared malnourished and some, in addition, to be dehydrated. All were over age 40. About 3 per cent were clinically obese, apparently on the basis of simple caloric excess, and all of these were between age 20 and HO. One suspicious case of ichthyosis and one of rickets were seen but in general there was no evidence of vitamin deficiency. Twelve cases (3 per cent) had clinical evidence suggestive of iron deficiency anemia. Three were pregnant women, the other 9 children under age 10. Two of the 9 children had positive stools for hookworm, one for Trichuris trichiura and one for Giardia lamblia. Several suspicious cases of venereal disease were seen but only one proven, this being a case of gonorrhea. -[- Several benign neoplasms, including two cases each of pigmented and nonpigmented nevi were found but no malignant tumors. Leukoplakia buc- calis was found in 14 natives, this lower incidence as compared to some other islands apparently being related to the fact that betel nut chewing is not a habit. Two cases of Mongoloid idiocy were discovered on Arno in brothers aged 8 and 24 years. No information could be obtained about the parents or other relatives. Eye, ear, nose and throat abnormalities were common. Pterygia were found in 71 persons (probably because of conjunctival irritation from coral) and chronic tonsillitis in 130 natives. ‘Three cases of bronchial asthma constituted the only allergic dis- orders encountered. Nine natives were suspected of having hypertension on the basis of blood pressures above the level of 140/90 on the single examination. No evidence of cerebral, cardiac, or renal abnormality was noted in conjunc- tion with these elevations. Eleven had external hemorrhoids and there were 7 natives with indi- rect inguinal hernia, two with umbilical hernias, and two with pilonidal cysts. Seven women had uncomplicated pregnancies. One woman age 60, a nullipara, had a complete uterine prolapse. Three cases with talipes equinovarus and one with polydactylitis constituted the major congenital orthopedic abnormalities. Summary The inhabitants of Arno Atoll appeared to be generally in excellent health and a staze of good nutrition. Tuberculosis appeared to be the single major disease tireatening health. ATOLL RESEARCH BULLETIN No. 63 Report on a visit to the Chesterfield Islands, September 1957 by F. Cohic Issued by THE PACIFIC SCIENCE BOARD National Academy of Sciences--National Research Council Washington, D. C. May 15, 1959 SpueTs[ Pletjste7zseyo SUTMOYS Bate ITJTOVeg 7SeMyANOS $ 3INOG31V3 6 ALIVAO1 . il? N e $30 a Wa3H $9))2N “Q e € » ti 01314431S 3H) $3]] JONVIY! Ati = SII ULSnV Report on a visit to the Chesterfield Islands, September 19572/ by F. Cohic Introduction The most westerly French possessions in the South Pacific, the Ches- terfield Islands, are of interest to zoo- and phyto-geography. They are at the junction of several biogeographic regions, Australian, New Cale- donian and New-Hebridean. There was a complete lack of information on these islands, so it was obvious that a visit would be fruitful. The visit, unfortunately very brief, only 4 hours, was made possible by the invaluable cooperation of the French Navy. We wish to express our grati- tude for the opportunity to the authorities in charge, particularly to Admiral de Toulouse-Lautrec, and to thank the officers and crew of the sloop Francis Garnier, who were very helpful to us. Location and history The Chesterfield group is located about 500 km. north-west of New Caledonia, in the middle of the Coral Sea, and stretches between 158 and 159 degrees of Longitude East, and 19 and 20 degress Latitude South. They include 11 islets, the main ones being the Chesterfield islets, Longue Island, and Avon, Renard and Bompton islets. Their total area is less than 1000 hectares. This coral group, halfway between Australia and New Caledonia, is completely uninhabited and isolated from shipping routes, and avoided as a danger to navigation. The Chesterfield Islands have been a French possession since 1876, and have only been rarely visited by whalers and warships. The last French vessel was the Dumont d'Urville which surveyed the island and prepared a map in 1939. Longue Island This is the principal islet on the V-shaped coral reef, and was the only one visited. It is a narrow tongue of sand, approximately 1800 m. long and 130 m. wide, and only 7 or 8 m. above sea level. It seems to be formed of a slightly raised platform, which has later been subject to wind erosion. The various borings made down to about 3.5 m. have revealed a regular stratification of darker guano-bearing sand lfnis report is an abstract translation by M.-H. Sachet of a French paper, issued here by permission of the author. The original, entitled Rapport sur une mission effectuee aux Iles Chesterfield en Septembre 1957, 20 pp., Oct. 1957, was issued by the Institut Francais d'Oceanie, Noumea, New Caledonia, where the author is an entomologist. ae layers alternating with lighter layers apparently free of guano, which seems to indicate that the surface has been repeatedly buried under new deposits probably by typhoons. No boulders or large size coral debris were observed. Along the shore can be seen some slabs of sandstone+ similar to those seen in the Loyalty Islands, a fact which may support the idea that the island is raised. From the point of view of pedology, the soil is made up of calcareous sands of a rather coarse texture, with a thin brownish humus-bearing layer, often buried under sand. The island has no fresh water. Vegetation Several plant associations can be recognized: a.) On the sandy, very sunny east slope Triumfetta procumbens forms large _ patches. Often associated with it, and even mixed with it Boerhavia repens (= Boerhavia diffusa) forms similar patches. Other plants found in this vegetation type are rounded patches of Coronopus integrifolius, a few clumps of Portulaca lutea and many scattered plants of Didiscus sp. (Umbelliferae). Cassytha filiformis is also associated with this vegeta- tion type, covering plants of Triumfetta and Boerhavia with its string- like stems. b.) On the east edge of the plateau the plants listed above are found again, but in smaller numbers. The low vegetation is made up mostly of grasses with Lepturus repens and Stenotaphrum sp. as dominants, forming a dense harsh turf, more or less continuous. There are only some clumps of woody plants, mostly in pure stands. The species are Sophora tomen- tosa, Scaevola sericea and Colubrina asiatica. The shrubs, particularly Sophora and Scaevola, are the principal nesting sites of frigate birds and certain boobies and thus determine the areas where guano is deposited. This vegetation is mostly dead with skeleton-like branches carrying the few twigs forming each of the numer- ous nests. The accumulated droppings burn the leaves and branches and perhaps the excess of toxic salts in the soil disturbs the physiology of the plants which are slowly disappearing. By contrast some isolated clumps, or some shrubs too low for the gregarious birds to nest on are remarkably luxuriant. c.) On the plateau, the vegetation is mostly made up of grasses of at least 5 species among which are Lepturus repens, Stenotaphrum sp., and Cynodon dactylon. The Lepturus cover is very hummocky in the central part, with deep hollows between the rounded clumps. Cynodon, or Bermuda grass, is very local and forms patches of very fine turf. Numerous clumps of Achyranthes aspera are scattered here and there, with occasional plants of a yellow-flowered Malvaceae, Abutilon molle, and of tobacco, Nicotiana tabacum. On the west part of the plateau, the beach morning-glory, Ipomoea, pes-caprae, forms a few patches. the photographs kindly sent by the author show this sandstone is typical beach rock. Ed. MEM nM Me ee MET a ks Gall ae : \ ee \ . 1. Longue \. CG SI aN Ilets du Mouilieg ty a "8 Ca Ilot du Passage 8, Pro RECIFS ET ILES CHESTERFIELD Echelle t_3_km. } 3 i 19°52° ILES CHESTERFIELD MOUILLAGE DE LILE LONGUE Echelle (490m. J} 13 158 /*20’ -3- The island also carries a dozen coconut palms which are growing rather poorly because of the violent winds, but particularly because of the bird droppings, which burn the fronds, and perhaps because of an excess of phosphate and nitrates in the soil. Flora The flora of Longue Island is remarkably poor. All the species pres- ent are strand plants of wide Indo-Pacific distribution, which are also found on the coral islets around New Caledonia. List of plants2/ of Longue Island ANGIOSPERMAE Monocotyledonae Gramineae Cynodon dactylon (L.) Persoon Lepturus repens (Forst.) R. Br. Stenotaphrum sp. Spinifex littoreus Thuarea involuta (Forst.) R. Br. Palmae Cocos nucifera L. Dicotyledonae Amaranthaceae Achyranthes aspera L. Nyctaginaceae Boerhavia repens L. Portulacaceae Portulaca lutea Forst. Lauraceae Cassytha filiformis L. Cruciferae Coronopus integrifolius Spreng. Leguminosae Sophora tomentosa L. Zygophyllaceae Tribulus cistoides L. 1/ Nomenclature and order adjusted somewhat from that in original French version. Kd. wihe i | . q Rhamnaceae Colubrina asiatica (L.) Brongn. Tiliaceae Triumfetta procumbens Forst. Malvaceae Abutilon molle Sweet Umbelliferae Didiscus sp. (probably D. cussonii Montronzier) Convolvulaceae Ipomoea pes-caprae (L. ) Roth Solanaceae Nicotiana tabacum L. Goodeniaceae . Scaevola sericea Vahl This flora of 20 species in 20 genera and 16 families is very poor. All the species are characteristic of coral strands; all are higher plants, no fern, moss or lichen were found. They are well adapted to the habitat and resistant to salt. Many have extensive root systems, and fleshy, or very hairy leaves. Most of them have fruits or seeds adapted to dissen- ination by ocean currents (corky fruits, long viability) or by sea birds (sticky or spiny fruits). Because of the extreme isolation of these islands, few species have been able to reach them and establish themselves. During our short visit to Longue Island, a few plants were intro- duced by the Service d'Agriculture of New Caledonia. These were: coco- nuts, Araucaria columnaris (Forst.) Hook., Sesbania erenuiiore Pers. and Leucaena glauca (L.) Benth. Terrestrial fauna The land fauna of Longue Island is very small. There are no mammals, | amphibians or land reptiles. The few groups represented are birds, tur- | tles - which come on land to lay eggs, sea snakes, a few Arachnida and insects. A, AVES ? The Chesterfield Islands, and particularly Longue Island, are an important roosting and nesting site for sea birds, which explains the formation of guano deposits. At the time of our visit the principal families represented were Sulidae and Fregatidae. Sulidae Sula sula rubripes Gould. The red-footed boobies nest mostly in the shrubby vegetation of Sophora tomentosa and Scaevola sericea. The nests contain only one egg, of a greenish white color and covered with a chalky deposit. -5- Sula dactylatra personata Gould. The blue-faced boobies nest di- rectly on the ground, without any twigs or nesting material, among the clumps of Triumfetta, Boerhavia and Lepturus. The nests contain two eggs but usually only one young. Sula leucogaster Boddaert. The brown boobies also nest on the ground but with a rudimentary nest of twigs usually protected from the sun. The females lay two or three eggs, but usually only one young grows up. Fregatidae | Fregata ariel Gray. The lesser frigate bird characterized by the two white lateral markings of the male is the most abundant. The nest is usually perched on Sophora tomentosa shrubs and built of twigs more or less cemented by a hard substance. It contains only one chalky white egg. Fregata minor palmerstoni. The great frigate bird appears to prefer Scaevola sericea bushes, but sometimes also nests on grass clumps in the open. Other seabirds of the family Laridae (Larinae and Sterninae) were li seen nesting on these low islands, but could not be identified. | B. CRUSTACEA 1. Isopoda-Oniscoidea At least two species are present and especially abundant under coral boulders and plant debris. 2. Decapoda-Anomura Coenobitidae Coenobita perlata M.-E. This hermit crab is one of the most obvious elements of the fauna of the Chesterfield Islands. This species of the tropical Indo-Pacific area is adapted perfectly well to terrestrial life and practically never goes to the sea except for egg laying. These ani- mals are of a beautiful red color, with little yellow tubercles bearing a black spine, and live on the part of the beaches above the reach of the highest tides, and also quite far in the interior, as we had already seen in Makatea. In the Chesterfield Islands, the hermit crabs are exceedingly numer- ous and can be found during the day under coral boulders, under drift wood (pieces of canoes, bamboos, tree-ferns) and plant debris. The speci- mens observed were all of a large size and lived mostly in shells of Trochus and Turbo. ‘They are mostly scavengers and clean the area of all sorts of material (fish, dead birds, etc.) but also eat the plants, as shown by their droppings which are full of plant cells and chlorophyll. However, while they are able to scale the coconut palms they do not ap- pear to attack them and are not responsible for their poor condition. + eee pe. Trommsdorff 1914 - Beitrag zur Kenntniss der in Deiitsch - Sudwest Africa verkommenden Zeckenarten; Arch. Schiffs. u. Tropenhyg., 18: 731-747. Van Zwaluwenberg (R.H.) 1955 - The insects and certain other arthropods of Canton Island. Atoll Research Bulletin, no. 2, p. 2. ATOLL RESEARCH BULLETIN No. 64 Canton Island, South Pacific (Resurvey of 1958) by Otto Degener and Isa Degener Issued by THE PACIFIC SCIENCE BOARD National Academy of Sciences--National Research Council Washington, D. C. May 15, 1959 Paves: Canton Island, South Pacific (Resurvey of 1958) by Otto Degener2/ and Isa Degener2/ The present paper is based on observations made on Cunton Island in February 1958, and resulting recommendations. It chiefly supple- ments observations made by Degener & Gillaspy on this atoll in 1950- 1951, and to get a better perspective of the situation the reader is advised first to study Atoll Research Bulletins 41-43 if he has not already done so, The present survey was made through arrangements with Mr, John M. Beardslee, Regional Administrator, Sixth Region, Civil Aeronautics Administration, Honolulu, Hawaii, aided by Canton residents Dr. Owen King and Mr. Eurl King. CLIMATE A table of meteorological data was published in Bulletin #41, pp. 7-8, taken from the U. S. Weather Bureau's publication, Local Climatological Data, ... Canton Island, South Pucific, for 1954. More recent editions of this Wcuther Bureau summary, through 1957, are available from the U. S, Superintendent of Documents, Washington ep, ), C.xfor d5¢eents. SOILS It is said that Canton was once so covered with guano that its harvest was long a lucrative export during the latter half of the Nineteenth Century. Even though most of the island has been disturbed by bulldozers, certainly some few areas must have escaped their turmoil as well as that made by the earlier shovels of the guano diggers, Yet, in company of the soil expert Dr. Lyle T. Alexander of the Soil Conservation Service, U. S. D. A., the writers could not find a speck of guano in February 1958. Has wetter weather leached guano out of the soil? Or has the coming of man so decimated the bird population that bird droppings now are washed apy by rain faster then: they can accumulate on the oma ; A feu sash groves of trees, growing in a 2 to 4 inch thick teat occur on Canton. Their probable origin began with a moderate guano | deposit stimulating herbaceous plant growth. This resulted in the formation and accumulation of decomposing organic matter.’. As-the soluble material washed out, a blackish soil high in phosphate re- mained suitable for the survival of trees. One of the soil samples collected yielded 11.4% phosphate on analysis. Perhaps it was such a phosphatic soil thet was exploited by the "guano" diggers, rather than true guano. VW New York Botanical Garden. 2/Formerly Botanisches Museum, Berlin-Dahlem. ee ee aos Borrowing Dr, Alexander's soil auger, we entered one Messerschmidia grove near the old guano wharf to take soil samples. This was near the lagoon. Here the humus-stained layer was four inches deep, an estimated accumulation of 1,000 years. The second place for samples was a Cordia grove near the British Settlement. This was near the ocean. The humic layer was shallower; the age, less. The samples, mailed to Dr. F, R, Fosberg, were assayed by the Geochemistry and Petrology Branch of the U. S. Geological Survey, Washington 25, D. C. The results are as follows: Sample 1 (Lab. No. 153526), top horizon in Messerschmidia grove, a dark brown sandy loam with some coral gravel that is somewhat weathered but internally clear white, Sample 4 (Lab, No. 153527), top horizon in Cordia grove, a dark brown loamy sand with light grains, very weak structure, Rapid rock analysis by Paul L, D. Elmore and Samuel D. Botts (methods similar to those described in U.S.G.S. Bulletin 1036-C. MgO CaO Na 20 K50 P20¢ Sample 1 ft eepeuroosy 26 AS .03 i? 2 Sample 4 2a0 44.6 AL .02 P72 Quantitative spectrographic analysis by Nola N. Sheffey (in percent on a moisture-free basis, Dried material was ignited at 900° C for 15 min.) % Loss on ignition Cu Mn Ni Sample 1 46.6 .0033 «0906 .0075 Sample 4 At. 3 .Q018 0902 3 3=9€,9002 Fe Cr ; sr Ba B Sample 1 «O13 0902 ie 002 O10 Sample 4 - O04 «.9002 50 Heo .008 Elements looked for but not found: Mo, Zm, Co. The above quantitative results have an overall accuracy of+ 15 percent except near limits of detection where only one digit is reported. In addition to these soil samples, foraminiferal sands were collected and the species identified. They will be reported on in a later paper together with other members of the animal kingdom. at LOCAL LANDMARKS The SS President Taylor, briefly mentioned in Bull. 41, p. 2, is a prominent, rusting hulk near Musick Light. It still dominates the wreckage-strewn British Settlement. Why the attempt to remove it failed is an interesting tale we owe to Mrs, Betty P. Defibaugh, former postmistress, The SS Granite State, 10,500 gross tons and an overall length of 522 feet, was launched in 1921. Two years later she was purchased by the Dollar Steamship Line and rechristened SS President Polk. She made about: a hundred round-the-world trips during the next five years with passengers and freight. Dec. 5, 1941, when in San Francisco, she was requisitioned by the Government to carry 200 military personnel to Manila, the planned sailing day being December 8, As the attack by Japan on Pearl Harbor intervened, plans were changed. The vessel was rechristened SS President Taylor, quickly remodeled, and sent to Honolulu December 27 with 600 officers and men of the U. S. Tanks Corps. She then returned to San Francisco with Army wives and children. February 3, with about 1,400 troops and equipment and accompanied by a destroyer, she sailed for Canton Island. As two Japanese sub- marines caught up with this little convoy near Canton, the President Taylor was ordered to steam close to the atoll. While the destroyer chased the submarines away, the ship ran aground, the troops disem- barking the following day, February 15. Various attempts were made to refloat her, the last one in May, without success. Canton being now an important war base, many of the troops sta- tioned there visited the wreck, helping themselves to whatever they could carry away to make their living quarters ashore more comfortable. The wreck itself remained a pleasant spot to visit until gutted by fire May 1948. The Caronia, a 300 ton steel nleasure yacht reputed to have cost 750,000 to build in 1927, was taken over by the Navy for the duration of the war, William Cooney, formerly in the Navy, and friends, finally purchased her for a few thousand dollars. They purchased the President Taylor for $5,200 at auction from the U. S. Maritime Commission. Incor- porated as "Taylor Salvors," the eleven partners with wives and child- ren to swell the total to 25, sailed on the Caronia for Canton with abundant salvage equipment and with food to last nine months, _ The salvagers cut off most of the ship above the waterline to lighten her, and patched the holes in the hull. Thereupon they planned towing her to Japan to sell for about 3250,000. But as an irreparable crack in the hull made refloating impossible, they had to be satisfied with salvaging as much metal as possible and selling it on Canton. Expenses just about ate up profits. Then real disaster struck. The night of July 11, 1956, the barking of dogs aboard the Caronia awakened the six sleepers just in time to enable them to jump overboard and swim ashore. The vessel was ablaze and soon a total loss. Salvage operation of the President Taylor ended, the last . . . 4 j Bhs partner forsaking Cunton January 1957. Then March 31 the Coast Guard towed the wreck Ceronia from the lagoon out to sea and sank her. The SS Granite State, alias President Polk and President Taylor still lies on the reef of Canton. The old Pan-American World Airways hotel, where the senior writer spent a few days in 1950, has been abandoned. It had become a shambles by 195%, Canton has improved since 1950-51 in respect to housing for CAA personnel, Many more attractive, modern homes have been built south of the Terminal Building. Some of these are surrounded by well kept gardens; a few by native vegetation and naturalized weeds. FLORA: SPERMATOPHYTA Bulletin 41 describes fully the native flowering plants, and lists the plant introductions made in 1950-51. Here we shall note pertinent observations about both categories, arranging the species in taxonomic order. Species which have not materially changed in status since the earlier survey will not be discussed here. Following instructions from CAA officials in 1950, most of the early introductions were made about the CAA housing area and Terminal Building. Though we were dismayed to see most of these plants with bare, dead stems in February 1958, we were gratified to notice some of them putting forth healthy shoots from the tase. The reason was not clear to us until a resident explained that this very area had been thor oughly sprayed; but instead of using the insecticide desired, the workaiey in error had used a weed killer! PAND ANACEAE Pandanus tectorius Park. of the Hawaiians is now represented by quite a number of specimens. All are restricted to gardens, and most probably stem from the 1950-51 introductions. GRAMINEAE Cenchrus echinatus R. Br. the sandbur, a terrible nuisance, is ubiquitous as before in areas where man is active. We mention it here because of the strange fact that it has hardly if at all increased its range since 1950-51. Near the plant nursery, now demolished west of the Terminal Building (Bull. 41, p. 39), two plants are killing and displacing this painful weed. Pennisetum setosum, because of its perennial habit, pree mpts areas where the annual sandbur dies of age, and takes over the ground before the germinated, bur-enclosed seeds can establish themselves, The other beneficial plant is the teaselgourd vine (Cucumis dipsaceus), which simply grows over the sandbur and smothers it to death. aie Chloris inflata Link, a fingergrass, conspicuously extends over neigh- boring vegetation. It is common about the old plant nursery, north of the "native" village inhabited by Gilbert and Ellice Islanders, and about the British Settlement near Musick Light. Cynodon dactylon (L.) Pers., Bermuda grass, thrives in several patches near the Terminal Building. The beautiful, green lawn existing about the home of a Hawaiian family in 1950 is gone as is its careful tender. He watered it from a brackish well he had sunk in his own garden. Digitaria henryi Rendle, a 1950-51 introduction, is established and forms small mats in the shade of the coconut trees about the guest house near the lagoon, British Settlement (Deg. & Deg. 24,647)... Digitaria pacifica Stapf, the native bunch-grass, is variable, as men- tioned in the previous bulletin. Along the lagoon, north of the native village, is a form (Deg. & Deg. 24,638) that extensively creeps to build up a huge, loose clump about 50 feet across. At the east end of the north Landing Strip, on the other hand, grows a very compact, dwarf specimen. Residents should keep such aberrant plants under observation to determine whether they are merely ecological forms or, instead, genetic ones worth describing. Digitaria sanguinalis (L.) Scop., crabgrass, if this difficult grass is correctly identified, grows here and there about the nursery area, It is reseeding itself. Digitaria timorensis (Kunth) Bal. is still growing in the same areas as in 1950-51. Now, however, it is likewise naturalized in dense but localized patches in the plant nursery area (Deg. & Deg. 24,645). Because of its perennial nature, it appears to be crowding out sandbur., Eragrostis amabilis (L.) W. & A., the feather lovegrass, is a delicate annual thriving here and there about the British Settlement. In the garden of the Terminal Building it forms a dense, tangled lawn, tending to crowd out all other herbs. Just before our departure from Canton in late February 1958, we were amazed to see all this grass being up- rooted and hauled away to leave a glaring white surface of coral shingle. Such an area becomes ideal for sandbur to take over, This little garden spot, seen by every voyager, needs a two to three inch covering of screened soi1,still quite plentiful near the old guano wharf. Then the area should be mowed regularly every month by a light hand lawn mower, whether it appears to need it or not, the gardener being careful not to let the wheels of the machine cut into the soil. Taller herbs will thus be discouraged from growth and, if the original workmen in February 1958 were not too efficient in eradicating every trace of it, perhaps the lovegrass will establish itself again from random seeds left in the neighborhood. Eragrostis whitneyi Fosb., the lovegrass named for my young friend, _ the late Leo Whitney, has perfected a means of racial survival in.this sun-scorched atoll by having shortened its life cycle to a scant three or even two weeks. February 1 we discovered a single tiny plant be- ginning to flower at the swimming hole near the lagoon, northwest side. Yee tee. see S. e e ! an ee a Ge Ey February 16 the species was very common locally, also actually on the Fighter Strip where no other flowering plants could gain a foothold, and here and there on the road at the southeast side, In the latter place it was among blue-green algae (Microcoleus acutissima and a little Schizothrix longiarticulata) that were binding the dusty sand together, This native grass does not grow in rubble, like lepturus, but in firmly compacted sand and coral fill. Perhaps in such terrain, soil moisture can more easily rise by capillary action to where the Shallow roots are located. The life span of this grass is even too short for the scale insect (Antonina graminis (Maskell) det. Miss Amy Suehiro) to grow to maturity about its rootstock. Lepturus pilgerianus Hansen & Potztal (1954) is of special interest, It is the only annual in the genus. Fosberg states, "The supposed annual habit of L. pilgerianus is especially meaningless to any one who has seen L, repens extensively in the field, as this feature is a purely facultative one, depending on the climate at certain stages in the life of the plants." This is a bold statement, particularly as the critic admits not having seen the type specimen. O. Degener spent a total of ten weeks botanizing on Canton; I.Degener spent three, The Degeners here reaffirm their previous finding that L. pilgerianus is annual. Furthermore, this annual plant differs from the perennial Lepturus growing with it in having culms red-violet not. green toward the base, in forming a narrower and less congested tussock or "bunch," in having leaves smooth rather than rough, in lacing obviously the dead culms and leaves of previous years' growth found in the perennial, and in the inflorescences at a distance appearing more conspicuously separate from the rest of the tussock. L. pilgerianus grows on both sides of the lagoon, being especially abundant near the Fighter Strip. Though passage was available for the asking on a Government vessel to neigh- boring Enderbury Island in February, due to an oversight the writers missed the opportynity of going there to ascertain the possible presence of L. pilgerianus. Iepturus repens (Forst. f.) R. Br., though recorded from Canton in prac- tically all previous papers, actually does not occur there in the narrow sense (var. repens) . The Canton plant is either the variety subulatus, originally described from Ujelang, Marshall Islands and widely distri- buted in the Pacific, or a variety still undescribed.:. The Ujelang plant "tends to be stoloniferous," a tendency foreign to the Canton plant. Panicum distachyum L. (Deg. & Deg. 24,649) is a new record for the iene” It forms dense mats in the British Settlement. One large clump (Deg. & Deg. 24,656) grows against the ruins of the PAA hotel, in a spot that receives rainwater from the roof. Panicum miliaceum L., growing near the PAA hotel in 1950 and atin: i to be derived from spilled ccs has disappeared. Pennisetum setosum (Swartz) L. Rich., introduced in 1950-51, covers an acre or two near the plant nursery to the exclusion of almost every other plant. Clumps occur sporadically elsewhere in the general neighborhood, Largely because of its perennial habit, as already aes | | mentioned under Cenchrus, it crowds out the annual sandbur. Like so many other introductions, Pennisetum has not extended its range extensively. These writers therefore in 1958 gathered many seeds, better termed grains, and scattered them in likely places, This grass and Cucumis dipsaceus (p 16) vie among themselves as the most valuable intro- duced sand binders, Setaria verticillata (L.) Beauv., the bristly foxtail was, next to the sandbur, the most troublesome weed in 1950-51. It was everywhere near Ny human activity, the coarsely hairy, fruiting inflorescences tangling I] themselves in socks and stockings of anyone straying from the center of a pavement. But during a three weeks' stay on Canton in February 1958 the writers found only two or three clumps. One of these was about a box of garbage that had been dumped on the roadside northwest of Turtle Beach. We do not know why this annoying weed has suddenly almost disappeared. Tricholaena rosea Nees, Natal grass, is now naturalized widely in the nursery region, about where it was first introduced in 1950-51. Cyperus javanicus Houtt., introduced in 1950-51, grows well in clumps Mh between the nursery and the ocean (Deg. & Deg. 24,650). Cyperus pQlystachyos Rottb. (Deg, & Deg. 24,655) grows in several clumps in the shade of coconut trees near the lagoon about the British Settle- ment. It is a welcome addition to the local flora, not recorded before. Cyperus rotundus L., the nutgrass, still grows about the old PAA hotel site; now also about the Terminal Building. no me menace theater of the time, south of the British Settlement; and Ff. pyenocephala Hbd. growing on a barren plain near the CAA housing area of the same time, are still there. By 1958 they had spread to the northwestern part of the. atoll, conspicuous and valuable sand binders of open areas everywhere, The latter resembles a coarse pincushion. The two species often grow together, and we noticed evidence of hybridization. ARECACEAE (PAIMAE) Cocos nucifera L., the coconut, now grows abundantly in gardens. Many are given a little care and hence are thriving. Only one tree, however, is now old enough to begin bearing. The healthiest trees, even though infested with scale, are in Dr. King's garden. They had been set in pani or depressions below the surface of the ground, holes dug deep enough to penetrate the hardpan. They were surrounded by large slabs of limestone tipped so as to lead an unusual supply of water to them at every rain, and were occasionally supplied with a little water left over after household use. This last treat will be hardly necessary after the roots have penetrated to near the brackish ground water table. a: Phoenix dactylifera L,, the date palm, was considered such a valuable food nlant in case of emergency on Canton that thousands of its seeds were scattered in 1950. Many germinated and Dr. S. G. Ross set outa score or so seedlings about his hospital in 1951 south of the British Settlement. In 1958 only two small date palms were growing on Seal 3 | cultivated in a garden. __ COMMELINACEAE Rhoeo spathacca (Sw.) Stearn, the white-flowered tradescantia or “oyster plant," was introduced in 1950-51. It grows well in some gardens. . Setcreasea purpurea B, K. Boom is cultivated about one home, and doing well. LILIACEAE Sansevieria cylindrica Bojer and its relative S. guineensis (Jacq.) Willd. ; introduced in 1950-51 are planted in gardens. None has become natura- lized. AMARYLLIDACEAE Crinum asiaticum L., or a species thus called in Hawaii and elsewhere, is planted commonly about homes, It is often grown crowded together, although it needs to be thinned to individual plants to get best results. Furcraea foetida (L.) Haworth, incorrectly listed as F, gigantea in Bull. 41 and by botanists in general, was introduced in 1950-51. It is grow- ing in a few gardens. MUSACEAE Mysa nana Lour., the Chinese banana, was cultivated in a few ao in 1958 to the exclusion of other kinds also ae in 1950-51. q 7 CAS UARINACEAER Gasherine eqiisetifolia L, and C. glauca Sieb. are both thriving without care, the latter increasing its area slightly by producing from its spreading roots adventitious buds that grow into new trees. -POLYGONACEAE Coccoloba uvifera (I.) L., the seagrape, was already growing and fruiting about the PAA hotel in 1950. But as Degener had a thriving tree in his garden at Mokuleia Beuch, Oahu, he gathered the abundant seeds fallen on clean sand, “his was more practicable than se:rching for and gathering the few that fell emong sandbur and foxtail under the Canton Atoll trees. The Oahu seeds were planted on Canton. By 1958, the old original sea- grape trees had formed a beautiful, small shady grove about the hotel grounds. Ripening fruit was everywhere, and "native" urchins were gathering them into tin cans as a source of food, The trees were reseeding them- selves and can furnish sufficient seedlings to form shaded roadsides aoe and gardens throughout the atoll if given a minimum of care. Volunteer seedlings, if left under the old trees die within a few years. Seed scattered about by Degener in 1951 had grown into a few low, spreading bushes about the bird refuge, and those planted south of the housi ng area were healthy and thriving. A few were planted at the Gillaspy home and, according to reports, were six feet tall in 1958 before they were uprooted and discarded by a new resident. AMARANTHACEAE Amaranthus dubius Mart. was not observed during our 1958 stay. ATZOACEAR Sesuvium portulacastrum var. griseum, first described by Degener & Fosberg as new from Canton, is still growing in the same areas as before, But in 1958 at the Dock a patch of the species, var. portulacastrun. (Deg. & Deg. 24, 651} was found naturalized. It is obviously of recent accidental introduction, seeds presumably reaching this spot with ship- ping. The difference between the slender, red-stemmed, red-flowered, glossy, terete-leaved var. portulacastrum and the stocky, yellow- stemmed, pale-flowering, dull, angular-leaved variety griseum is striking when the two grow side by side. PORTULACACEAR Portulaca cyanosperma Egler, a Hawaiian purslane with pretty violet- red flowers, was introduced in 1950-51. Mr. & Mrs, Albert Lincoln reported the former occurrence of large mats of this succulent near Turtle Beach, northeast side of Canton. They transplanted some to their garden near the lagoon where the writers happened to see it and its escaping offspring. A visit to the former Turtle Beach stand dis- closed that waves of a recent, severe storm had washed over this area, Sweeping all vegetation in their path, except Suriana, into the lagoon. CARYOPHYLLACEAE Spergularia marina (L.) Griseb,, a pale, fleshy halophyte (Deg. & Deg. 24,612), was collected on Johnston Island January 30, 1958. Its seeds were scattered about the new hospital in the CAA housing area on Canton a few days later, CASSYTHACEAE Cassytha filiformis L., the love-vine, is native to Canton. It now covers like a huge greenish throw-net several acres near the Shark Pool, parasitizing Portulaca lutea, Boerhavia and Sida. CRUCIFERAE Lepidium o-waihiense Chau. & Schlecht., introduced from Oahu in 1950-51, is surprisingly abundant (Deg. & Deg. 24,636) and even crowded in 1958 on part of the elevated side of the causeway south of the "native" village, The writers gathered several handfuls of seeds to scatter in other likely places. ; Seth O)ews CRASSULACEAE Bryophyllum pinnatum (lam.) Kurz, the air-plant so famous for preducing plantlets from notches in its leaf even when the latter is pinned on a window curtain, is now growing in a few gardens. It has not escaped, but should be encouraged to do so, LEGUMINOSAE Desmanthus virgatus (L.) Willd., the slender acuan, was introduced in 1950-51. Numerous branches of this shrub stand stiffly upright about the nursery area, killed by the weed spray. ILuckily two or three root- stocks survived the poisoh, and in February 1958 bore flowering branches. Leucaena glauca (L.) Benth. (haole koa in Hawaii) was already represen- ted by a few fruiting plants in 1950 at the PAA hotel entrance. Dege-~ ner in 1950-51 introduced many seeds. Now this plant is sparsely re- seeding itself about the British Settlement, and fairly well in the nursery area, even though recently exposed to the weed spray. Prosopis chilensis (Molina) Stuntz, the algaroba, is a shallow-rooted tree related to the deep-rooted mesquite of the southwestern part of the United States and northwestern Mexico, As stated previously (Bull. 41, p. 33), the Degener-Gillaspy policy was to introduce no plants that might be poisonous or painfully thorny "such as the... algaroba (Prosopis chilensis)", Nevertheless, in 1958 we were pleased to see a dainty algaroba growing in a flower pot on a resi- dent's porch. There is a vast difference between algarobas growing under proper confinement and such trees growing wild to scatter their cruelly efficient thorns where such a large proportion of the popula- tion goes barefoot, EUPHORBIACEAE Acalypha wilkesiana M, -A,, the painted copperleaf, had disappeared by 1958, but the three Chamaesyce species were as common as before. Tney had not materially extended their range. Phyllanthus "niruri", "growing in 1951 in and about a box of soil im- ported from Fiji, pal “had established itself about the British Settle- ment, but was rare in 1958, When this specimen (Deg. & Deg. 24,646) is endicd further it will provably be shown to be another species, as according to G, L, Webster (1956) P. niruri L. though widely re- ported; does not occur in the Pacific Islands. Potnsettia cyathovhora (Murr.) Kl. & Garcke), the fiddle-leaved poin- settia, in in 1950 was sparingly naturalized and protected in the British Settlement; but not observed elsewhere. In 1958 this modest ornamental was growing wild and common in the housing erea. When the plants flower, thirsty flies and other insects are attracted to the nectar. Some housewives therefore keep the plants from growing near their homes, The garden poinsettia (P. pulcherrima (Willd.) Graham) was no- where to be seen. It is worth introducing again as a garden plant. il ANACARDIACEAE Schinus terebinthifolius Raddi, the Christmasberry tree, is represented by a small healthy plant in a garden near the airport. MALVACEAB Gossypium brasiliense Macf., the kidney-cotton of commerce, was intro- duced in 1950-51 about the CAA housing area, Several years later it was abundant, and conspicuous with its untidily spilling seeds and white lint. After the disastrous spraying of the area with weed killer, this choice ground cover was killed except for a few score plants. These were producing their crop of cotton in 1958, and some of the seeds were planted at that time. The endemic Hawaiian cotton (Gj tomentosum), likewise introduced in 1950-51, is nowhere to be seen. Hibiscus rosa-sinensis L., the common Chinese hibiscus, is growing in a few gardens as before. The plants are chlorotic. Pariti tiliaceum (L.) Britton (hau in Hawaii) was already on Canton when Degener introduced many seeds from Oahu in 1950-51, In 1958 the tree was rarer than before; the old one at Musick Light still lives. Sida fallax Walp. is almost everywhere as before. Near the garbage dump south of the housing area grows a freak bush with almost white flowers. Thespesia populnea (L.) Sol. (milo in Hawaii) still grows on Canton. | In spite of the numerous seeds introduced in 1950-51, it is repre- sented by only a few more individuals in 1958. CLUSIACEAE (GUTTIFERAE) Calophylium inophyllum L. (kamani in Hawaii) was already on the island before Degener imported seeds and potted plants in 1950-51. The old trees of 1950 about the PAA hotel are growing well. TAMARICACEAR Tamarix aphylla Karst., the European tamarix, is represented by the same beautiful, large trees that were growing about the hotel in 1950. Cuttings from Oahu set out by Degener in 1950 about the Weather Sta- tion had been accidentally killed by the weed killer except for one tree that has become ten feet tall. This represents a growth of al- most one foot per year. PASSIFLORACEAE Passiflora foetida L., the foetid passionflower, was represented by a single plant in 1950 before Degener introduced and scattered hundreds of seeds. In 1958 it was naturalized about the plant nursery and Brit ish Settlement. The long vines are flowering, bearing their edible fruit, and reseeding themselves, -12- CARICACE AR Carica pa paya L., the papaya, was planted as before without following the writer's suggestions of adding humus in the soil about it (Bull. 41, p.26). As exnected, the trees were chlorotic and non-fruiting. Conocarpus erecta L., introduced in both the glabrous and silvery vari- eties by Degener in 1950-51, are beautiful shrubby trees in the Terminal Building garden. The ground under the trees is covered with seedlings available for replanting, They have not been used and gradually die due to competition with the parent plants. Terminalia catappa L., the Indian almond, was already on Canton when Degener introduced seeds and seedlings in 1950-51. The Degener intro- ductions were nowhere to be seen in 1958, whereas the old trees of 1950 about the PAA hotel are not only growing well but reseeding them- selves, Young plants up to a foot high are under the old ones just - waiting to be transplanted, ARALIACEAE Polyscias guilfoylei (Bull) L. H, Bailey, the panax of Hawaiian resi- dents,is now growing in a garden or two in the housing area. APOCYNACEAE Plumeria rubra L., the frangipani, was introduced in a cultivated form previous to 1950, and died that year due to scale insects, Offspring of wild trees originally gathered in the Bahamas by Degener and planted by him on Canton in 1950-51 have disappeared. These bore small, white flowers, A single plant of a cultivated fo obviously recently introduced, now grows in a garden. Calotropis gigantea (L.) Ait. and its white-flowered form nee Degener, the crown flowers, are grown with success in both the British Settle- ment and the CAA residential area, As these milkweed are host for the caterpillar of the monarch butterfly, some residents destroy their plants rather than bother to kill the insect with a spray.or by hand picking. We are not certain whether the lavender crownflowers in gardens come from cuttings imported’ in 1953-51 or are from the ancient bush still growing between Musick Light and the wrecked "President Taylor." The variety wilderi, in the original description by the senior writer - in 1937, was stated to have been introduced into the Hawaiian Islands "by Gerrit Wilder "from Tahiti, where he found it, growing in the garden of Robert Keable, writer.'" More recent evidence indicates that Wilder erred in claiming he had first introduced this plant into Hawaii. It had been growing there years before his introduction. Whether it originated in the Hawaiian Islands, Tahiti or elsewhere is still an open question. qos CONVOLVULACEAE Ipomoea tuba, which was not identified with certainty in Bulletin 41, p. 27, is a correct scientific name for the native moonflower. This twiner is represented by the same old plants observed in 1950-51 along the northeastern rim, It has not spread of its own accord since that time. The few seeds lying about the dormant plants in 1950-51 were gathered and planted about the CAA housing area, It was refreshing to see this night-bloomer cover now almost an acre of rubble where the two landing Strips meet, and to see a patch a hundred feet wide cover- ing Sesuvium just south of the housing area. A few moonflowers are being successfully trained on trellises, The plant is beautiful at night with its large, gleaming white flowers facing the stars overhead, Ipomoea batatas (L.) Poir., the sweet potato, was observed in 1958 . planted in the open in the housing area. It thrives except for chlorosis. Ipomoea pes-caprae (L.) Roth, the beach morning-glory, frequently re- corded from Canton, is probably represented only by the subspecies brasiliensis. We do not know whether we should classify it as native or introduced; it is a borderline case, So far as its seeds drift to Canton and germinate there, it is a native. But due to the terrestrial hermit-crab (Coenobita perlatus),such plants never survive, We now believe the only plants able to mature receive some human protection. In this sense, the beach morning-glory must be classed as introduced. In February 1958, hundreds of thousands of beach morning-glory seeds were seen washed up in an almost solid line along the north shore of Canton. With several other species of various families, they germinated well enough to mark a thin greenish line along the beach. The charac- teristically shaped cotyledons were soon devoured and not a seedling was observed with normal, goat-footprint leaves. Yet about the hotel area and CAA housing area, where Degener planted Oahu seeds in 1950- 51, the vine was common, covering the ground with its long, flowering, erpeping stems. These even covered the bare floors of the large Quonset huts near the lagoon beach, Tuough a prostrate creeper, not a high twiner like the moonflower, the slender stems were successfully trained by some residents over fences near the Terminal Building to produce a truly charming effect. BORAGINACEAE old PAA hotel grounds. Its roots, like probably those of all the various kinds of older trees in the vicinity, may have reached to near the watertable, There, one must remember, the water is but mild- ly brackish. The ground under the geigertrees is covered with healthy seedlings, material that could be nicely planted in gardens and along the roads of the CAA housing area. Cordia subcordata Iam. (kou in Hawaii) may be native according to some evidence; originally introduced by man according to other. For example, during our rather thorough search for drift material we failed to find a single kou fruit, notwithstanding that they are corky and hence ad- mirably suited for ocean transport. About the present kou groves, where hermitcrabs congregate mainly to enjoy the shade, the seeds Se epee = eae readily germinate but the writers have yet to find a single one escape crab destruction to grow to maturity. The opportunity for a rare seed during the atoll's life-time to reach a favorable spot, to be covered with sand and germinate, and finally to mature in spite of hungry crabs is entirely possible but unlikely. Furthermore, clusters of kou trees about the old guano wharf "are said to have been planted years ago (Bull. 41, p. 28)." That leaves the two groves, less than one mile south of Musick Light, to consider, The one covers about 3500 square feet and the other about 2700 (Bull. 43, p. 3). The soil in the former, according to Dr, Alexander's estimate to us on the spot, is several hundred years old. Do these old kou groves, each probably derived from a single fruit, represent an introduction several hundred years ago by ocean currents or by early Pacific Islanders, perhaps those who may have erected coral rock edifices (Bull. 41, p.6)? Even though the kou was already there, Degener shipped a large quantity of fruits from his Mokuleia Beach, Oahu, trees to Canton in 1950-51 to add to fruits gathered by him locally. Then while hunting for and studying specimens south of the housing area, he was wont to wander over the area, a shovel in the right hand, and a pail with vials, killing bottle, flask of water and kou fruits in his left. At various spots he thrust his shovel a few inches into the friable soil, moved it back and forth a few times, threw a small handful of fruits in the gaping hole, and tamped the soil about them. In 1958, the land- scape, thanks to this early planting, was dominated by vigorous clumps of kou trees, each about six feet tall and seven or eight years old. The very minimum care in planting combined with isolation, that made it difficult for crabs to find everyone of the thousands of tasty seedlings, was sufficient to establish groves of this tree. Defolia- tion by caterpillars, an ugly nuisance to which all kou are presently Subject, will be lessened thanks to Territorial Entomologist G. C. Chock of Honolulu. He is preparing to ship the proper parasite for control, VERBENACE AR Stachytarpheta jamaicensis Vahl, called oi in Hawaiian, is native to tropical America. The small quantity of seed available was scattered in 1951 mainly about Degener's cottage near the lagoon south of the PAA hotel, In 1958 the oi was locally abundant and naturalized. The only other station observed that year was at the garbage dump, south of the CAA housing area, This consisted of a single large plant, in that area decidedly ornamental with its pale lavender flowers. Be- cause of their contrast against a field of yellow Sida, its scanty Sprays were being picked for ornament. Its popularity may yet cause its death. At the abandoned ccttage, on the other hand, the plants can continue to blo:som and reseed themselves undisturbed. SOLANACEAE Capsicum frutescens L., the common redpepper often called "Chili pepper," is more than holding its own in a garden. ee Lycopersicum esculentum var. commune Bailey, the common tomato, was evi- dently recently introduced and thriving in a garden. L. esculentum subsp. galeni (Mill.) Buckwill, a wild tomato with small fruits naturalized on Oahu, was hopefully introduced in 1950-51. Not a single plant was to be seen in 1958, "Lycopersicon esculentum var.", collected near the Airport by Degener & Hatheway (No. 21,307) in 1950, was likewise gone. It re- sembles a plant from the Galapagos (Alban Stewart No. 3,380), and like it is a straight variant of L. esculentum, Nicotiana glauca R. Grab,, the tree tobacco, already in Canton in 1950, was reintroduced by Degener the following year. In 1958 one huge plant, a respectable tree, grew in a depression north of the north landing Strip close to the ocean. It had been badly mauled by a storm, but was re- cuperating well from its injuries. It had reseeded itself prolifically during the past years, and the younger plants, likewise injured during the same storm, were likewise growing lustily. Other plants growing elsewhere often become chlorotic, Petunia hybrida Nilm, the common garden petunia, grows nicely in a win- dow-box. Such prized plants almost always receive a little care, some water left over after rinsing a vegetable or the dishes, perhaps even a little commercial fertilizer, Such ornamentals are well worth grow- ing, but hardly can be expected to escape into the wide open spaces where such little extra care is denied them. Physalis angulata L., the husk tomato, is found, as in 1950, sparingly naturalized at the British Settlement. This time it has spread to the Terminal Building garden. Here the plants suffer from the attacks of red sp:iders, Solanum melongenasman esculentum Nees, the eggplant, grows in a garden. It promises a crop. RUBIACEAE Casasia clusiifolia (Jacq.) Urban, perhaps best named the Bahama gardenia, was introduced as small plants in 1950-51. Seeds, collected by Degener many years ago along the arid coral coast of New Providence, British West Indies, were sent to Colin Potter, custodian of Foster Botanical Gardens, Honolulu, for cultivation and distribution. In 1951 Mr. Potter kindly donated some healthy, six-inch plants for Canton, which were forthwith set out at the Terminal Building. In 1958, though somewhat infested with scale insects, the Bahama gardenias had deve- loped into beautiful compact shrubs, three.to four feet high, with dark, glossy leaves and sweetly fragrant, gardenia-like flowers. Even the large orange fruit, edible in an emergency, was setting but we do not know whether its seeds will be viable. This ornamental ever- green is highly recommended for planting in local gardens. Morinda Gitrifolia L.,(noni in Hawaii) was known to Degener in 1950 by a Single large tree near the old guano wharf, It had disappeared by 1958. At that time, however, a large, but dilapidated tree was growing in the "native" village and, according to a Gilbertese, several more were growing east of the north Landing Strip. eThe variegated var. potteri Degener, introduced by Degener originally from Viti Levu, Fiji, planted in 1950-51, was nowhere observed. CUCURBITACEAE | Citrullus vulgaris Schrad., the watermelon, was planted in one garden. Cucumis dipsaceus Ehrenb., the teaselgourd of Arabia, was introduced in 1950-51, At the old nursery in 1958 it had covered one area 50 feet across, and was growing in several more places in the vicinity, In the former locality this vine formed a Juxuriant, densely leafy ground-cover, smothering out every other plant in the area, including the obnoxious sandbur (p. 4 ). The ground was covered with golden yellow gourds, many of which were gathered, sliced and scattered in likely, places. Though this interesting plant finds Canton to its liking, it seems not to spread much farther than its slender extending vines can drop their seed-filled gourds upon the ground. An efficient animal for proper seed-dispersal is wanting.. Grown for ornament and curiosity in many lands, it is strange no island resi- dent has trained the teaselgourd over fences and trelisses like the moonflower and the less adapted beach morning-glory. Cucumis melo L., the muskmelon, thrives in one garden, the vines covered with unripe melons two to three inches in diameter. Even if they do not ripen, they can be made into a conserve or pickle. According to Mr, Beardslee, many ripened, ‘GOODEN TACEAE Scaevola, the naupaka of Hawaiians, consisted of two kinds of plants on Canton in 1958: the native one, common almost everywhere; and the Hawaiian one, consisting of less than a dozen plants derived from Cahu seeds scattered in 1950-51. To have these two kinds growing to- gether under identical conditions was ideal for their comparison and study. The Canton plant becomes ten feet tall; has glossy, hairless leaves; and white flowers in which the corolla lobes are marginally narrow-winged and slightly fimbriate. The Hawaiian plant, on the contrary, rarely attains half that height; has finely velvety leaves; and white to more or less lavender flowers in which the corolla lobes are marginally broader-winged and more prominently fimbriate. Residents of Canton, without my botanical knowledge, had noticed the difference between the two even before our arrival from Honolulu. The coastal species of Scaevola on Canton and other Pacific Islands has white fruit and has most often been called S. frutescens (Mill. ) Krause. Fosberg (1956) has demonstrated that this name can be applied only to the Atlantic coast Scaevola with black fruit, which the writer knows well from Bermuda and the Bahamas. Fosberg concluded that the only name available for the Pacific species is Scaevola sericea. Vahl (Symb, Bot. 2: 37, 1791). This was described from Niue (Savage Island) east of Tonga and the name can best be trans- lated as silky scaevola. While the Canton plant is nowhere silky, we tentatively call it S. sericea until further studies determine whether another name is necessary. ow Jas ‘The common Hawaiian plant has been called variously S. lobelia, S, frutescens, S. frutescens var. sericea and S. fauriei. We know it cannot bear any of the three first names. We likewise believe it is not sufficiently distinct from the Niue plant to have a specific name of its own, We therefore (Phytologia 6(6): 321, 1958) named the Hawaiian plant Scaevola sericea var. fauriei Degener & Degener. Now that two related naupaka grow on Canton, it will be interesting to note whether they will hybridize. COMPOSITAE Conyza canadensis (L.) Cronq. (Deg. & Deg. 24,961) is represented by about a dozen plants. These grow about a warehouse southwest of the Terminal Building, mostly under the eaves of the roof, perhaps because of the additional water that occasionally drips from them. This plant is new for Canton, an accidental introduction. Gaillardia picta Sweet, a variable form of hybrid origin naturalized on dunes about Laie, Oahu, was sown around the Terminal Building in 1950. In 1951 several plants had already flowered, most of them under the eaves of old shacks. In 1958 the shacks were gone, and the Gaillardia as well. With a little care, this ornamental would reward the gardener with an abundance of pretty flowers varying from yellow to orange-red. Pluchea indica (L.) Less. , the Indian pluchea, was introduced in 1950- 51. It is thriving and common about the Turning Basin. It now grows among its coarser, grayer relative P, odorata (L.) Cass. Tagetes sp., the marigold, was observed cultivated in a window box. if cvanopuyra®” Cyanophyta (Myxophyceae, Cyanophyceae) commonly called blue-green algae because of their prevailing color, are extremely important in keeping the sand and the coral dust of Canton Island from blowing away. Some "blue-greens" form a felt-like covering near the surface of the soil; some cover the soil with a gelatinous matrix; others evi- dently cambine with bacteria to form a thin layer of stone; while still others are apparently of no value at all as soil binders. When Canton suffers from an exceptionally long drought, some of the felt- like blue-green algae curl up in leaf-like flakes and eventually may erumble into dust, thus exposing the loose sand and soil as well as themselves to wind erosion. Of course a ship or plane may have unwittingly transported different kinds of blue-greens from one island to another. Perhaps because of such agency of man, Entophysalis deusta and Microcoleus ere groups of algae in the Degener Collection are described by E. Yale Dawson as Bull. 65. = ET, eet cit ase. OE PLL SS EE NE ey a a ia chthonoplastes were collected on both Canton Atoll and Johnston Island (Deg. & Deg. 24,580 and 24,580a, Jan. 30, 1958). The consensus of botanists, however, is that such microscopic organisms are readily trans- ported even by high air currents throughout the world. To be wafted from Canton to Johnston or vice-versa is not a difficult feat, one that probably occurred time and time again. Many blue-greens are capable of survival on sun-scorched rocks, as found on Canton, on the Equator; and in the thermal springs, as in Yellowstone and New Zealand... Such plants are likewise capable of survival and growth in frigid regions. Thanks to Dr. Francis Drouet's efficient advice on the collecting of Cyanophyta received before our February 1958 stay on Canton and thanks to his expert determinations, we have the following to report: CHROCCOCCACEAE Anacystis aeruginosa (Zanard.) Drouet & Daily (Deg. & Deg. 24,744). Anacystis dimidiata (Kitz.) Dr. & Daily (Deg. & Deg. 24,584, Fab. 16, 1958), with h Entophysalis deusta, was found along the Southeast beach of the lagoon in shallow water gently rippled by wave action. It imparted to the fine, almost muddy, white sand, a pale pea-green color.Mixed with Johanne HEELS pellucida and Huaakys aestuarii which see, Anacystis montana (Lightf.) Dr. & Daily (Deg. & Deg. 24,571, Feb. 3, 1958) occurred as a thin mat about one mile east of the Fighter Strip along the wet edge of a more or less pérmanent shallow, narrow pond. Various kinds of fish must have washed over the beach into.this pond during severe storms but we found only mullet. The reason they could Survive while the others did not was evident on tasting the water. it was almost fresh. Johannesbaptistia pellucida (Dick.) W. R. Taylor & Dr. (Deg. & Deg. 24,573, Feb. 8, 1958), commixed with Entophysalis deusta, Anacystis green mass at the bottom of a fresh-water pond behind the beach near the Messerschmidia woodland shown on Hatheway's map (Bull. 43) near. the center of the north shore. CHAMAES TPHONACEAE Entophysalis deusta (Menegh.) Dr. & Daily (Deg. & Deg. 24,594, Feb., 1958) is the microscopic plant that by its astroncmical numbers imparts to the atoll a dark grayish somberness excepting where displaced by’ the green of other vegetation, the gleaming pale yellow to pink of the beach sand, the white of wave-worn coral slabs, and the indescribable pale greens to aquamarine blues of lagoon and ocean. This so-called blue-green alga covers the upper surface of coral slabs, barely pene- trating beyond superficial interstices. Though useful in cutting down the glare of the otherwise alabaster-white coral rock, no one has yet worked out to what extent its absorption of the sun's rays raises the temperature of its surroundings. Considering the climate, these primitive plants are sun-scorched and dormant most of the time, springing Brie fi into life during occasional rains and ocean mists; or at night when much of the atoll is covered by a film of moisture readily formed on surfaces impregnated with ocean salts, This same plant, in conjunction with bacteria on and about decaying plant and animal debris, formed a firm 8 mm.. thick crust about a drying, brackish pool on Spam Island (Deg. & Deg. 2%, 593, Feb. 19, 1958). Even an arid undermined ledge of limestone overhanging the lagoon bore this persistent, difficult-to- kill plant on both its surfaces (Deg. & Deg. 24,568, Feb. 8, 1958). It (Deg. & Deg. 24,564, Feb. 14, 1958) even blackened the limestone bottom of a drying, shallow arm of the lagoon six miles east of the Fighter Strip. It is almost everywhere, and hence it would be repétitious to list all our collections, “To realize fully the frequency and island-wide distribution in various habitats of this species, it is merely necessary to peruse the para- graphs below concerning other species and to learn that E. deusta was growing with many of them. OSCILLATORIACEAR Hydrocoleum comoides (Harv.) Gom. (Deg. & Deg. 24,918, Feb. 1958) was found on the lagoon side east of the north Landing Strip. For another collection (Deg. & Deg. 24,576a, Feb. 3, 1958) see wmder H. glutinosum of same number, Hydrocoleum confluens (Setch, & Gardn.) Dr, (Deg. & Deg. 24,727, Feb. 2, 1958) grew on reef rocks, in shallow water at low tide, northwest of the Gilbert-Ellice Village. Hydrocoleum glutinosum (Ag.) Gom. (Deg. & Deg. 24,578, Feb. 2, 1958) stained patches of muddy sand a light green. This was common on the ocean reef, shallow at low tide, southwest of the Terminal Building. The same species (Deg. & Deg. 24,576, Feb. 3, 1958) and some H, comoides were collected about 10,000 feet east of the northwest tip of Canton, Here on the ocean side is a limestone reef so flat and smooth as to remind one of a concrete sidewalk. In many areas a very fine white sand with a light bluish green color covered this rock to a depth of about one centimeter. Instead of being washed away by the waves of the open ocean, this sand remained, probably held in place by the gelatinous secretion of H. glutinosum and the less abundant H. comoides. Hydrocoleum lyngbyaceum Kitz. (Deg. & Deg. 24,916, Feb. 18, 1958) grew on the ocean reef, south shore, in company with the former species; and (Deg. & Deg. 24,785, Feb. 6, 1958) with Spirulina tenerrima Kutz. lyngbya aestuarii (Mert.) Liebm. (Deg. & Deg. 24,577, Feb. 7, 1958) occurred for a depth of half a centimeter or less in the calcareous sandy mud exposed at low tide along the artificial channel southwest | of the Turning Basin. Another collection (Deg. & Deg. 24,583, Feb. 14, 1958) was made six miles east of the Fighter Strip in a fresh to brackish water puddle, slowly drying out. This Lyngbya, with an ad- mixture of Entophysalis deusta, covered the area with a dark bluish green, somewhat gelatinous mat a few millimeters thick, With Anacystis dimidiata, it (Deg. & Deg. 24,565, Feb. 18, 1958) formed a ee ar. > 0 Tl PS ES a es A Re NS ee wore very dark blue-green covering over the bottom of a slightly saline, shallow pool on the lagoon side of the southeastern tip of Canton. Regarding still another find of_L. aestuarii (Deg. & Deg. 24, aie Feb. 19, 1958) see Phormidium gardnerianum. Iynebya confervoides Ag... (Deg. & Deg. 24,770, Feb. 17, 1958) grows on old coral on the ocean reef, shallow at low tide. The spot is on the south side, two miles east of the Kou forest. L. majuscula (Dillw.) Harv. Lyngbya infixa Fremy (Deg. & Deg. 24,745) on lyngbya majuseula (Dillw.) Harv. (Deg. & Deg. 24,749). Iyngbya semiplena (Ag.) J. Ag. (Deg. & Deg. 24,575, Feb. 5, 1958) grew along the north shore, quite a distance inland from the beach. Though exposed to full sunlight, it bound the sand together for a thickness of three to five millimeters. A quite different habitat for this species (Deg. & Deg. 24,917, Feb. 8, 1958) was old coral forming the submerged part of the guano wharf in the lagoon, At the bird reserva- tion, on the other hand, it (Deg. & Deg. 2%,775, Feb. 16, 1958) formed a thin brown film on that part of the ocean reef that is partly exposed at low tide. Microcoleus acutissima Gardn. (Deg, & Deg. 24,,579, Feb, 14, 1958). with § Schizothrix : jongierticulata, grew on the flat bulldozed rubble surface of the ‘Fighter Strip, to some extent inhibiting the formation of dust clouds. In part of this area, among the alga, grew the native dwarf lovegrass, Eragrostis whitneyi we eneenincae * Microcoleus chthonoplastes (Fl. Dan.) Thur. (Deg. & Deg. 24,597, Feb. 13, 1958), with Entophysalis deusta, was collected at the southeast end of the lagoon. At low tide these extensive mudflats, here and there inhabited by colonies of gaudy fiddlercrabs (Uca tetragonon), are sun- baked and occasionally even entirely dried, Here one or both of these blue-green algae, perhaps in conjunction with: bacteria, form a stiff, brittle crust of rock about one millimeter thick. One mile west of Shark Pool, southeast tip of Canton, M. chthonoplastes (Deg. & Deg. 24,581, Feb. 15, 1958) formed a hard crust of the surface mud of a drying fresh-water puddle. For another collection: (Deg. & Deg. 24,,572a, Feb. 2, 1958), see Scytonema hofmannii. Microcoleus paludosus (Dezenen 21,341} reported in Bull. al p. 40, is a misidentificaiion for M. chthonoplastes, mentioned above. Microcolevs tenerrimus Gom. (Deg. & Deg. 24,595, Feb. 1, 1958). On the coral slabs cast up onto dry land by storm waves and there left to rest undisturbed for many months and even years, the dark Entophysalis deusta thrives, as described in detail under that species. Nearer the beach, however, are similar coral slabs, somewhat more often tumbled about by wave action and hence alabaster-white. Such flat stones, because of their coral origin, are translucent and of course more or less porous, the degree depending upon the precise species of animal. Due to the intensity and the drying effect of the sun, perhaps the upper nine-tenths or even more of the slab is relatively devoid of -21- plant life. But the remaining area, as in a greenhouse or fernery, supplies an environment of diffuse light and relatively constant mois- ture, particularly if lying on moist sea sand. For this reason it is a pleasant pale green. An occasional cuttle-bone, the inside shell of the cuttle-fish, lies among such coral, As it has the same general appearance, we suspected it to be a similar locus for the growth of Cyanophyta, Stone-like coral being difficult to handle, we sent the cuttle-bone to Dr. Drouet for dissection. Here he found Microcoleus tenerrimus, Plectonema terebrans, Entophysalis deusta and some Calo- thrix crustacea, Phormidium gardnerianum Dr. (Deg. & Deg. 24,592, Feb. 19, 1958), with Lyngbya aestuarii, colored spinach-green most of the bottom of a brackish pool on Spam Island. Phormidium papyraceum (Ag.) Gom. (Deg. & Deg. 24,592, Feb. 19, 1958), associated with Nodularia sphaerocarpa and Plectonema nostocorum, was collected about a mile east of the Landing Strip, northern Canton. Here the Armed Forces during the Second World War had constructed a cement trough three by six feet in area and a few feet deep. This was permanently rain-filled and here P, papyraceum formed dark, paper-thin, blue green pellicles more or less attached to the trough sides. Among this well-named alga lived the bloodred larva of the midge Tendipes esakii, originally described from Saipan. Porphyrosiphon fuscus Gom. (Deg. & Deg. 24,598a, Feb. 11, 1958), see Schizothrix lamyi. Schizothrix creswellii Harv. (Deg. & Deg. 24,574, Feb. 8, 1958). Tridaena shells are everywhere; but the living mollusks, standing up- right above the general surface of the water shallow at low tide, are rather localized, The shells are agape, exposing to the sun their voluminous mantles which are of strikingly beautiful pastel shades, almost gleaming in the sun. These vary from purples and dark browns to yellows, greens and blues. It is reliably reported that the unusual color of the mantle is caused by microscopie blue-green algae living in symbiotic, almost captive, relationship, with the tridacna. We killed no tridaena for samples of blue-greens; living tridacna we knew only from the ocean reef. Then Mr. & Mrs, Albert Lincoln brought us to an unusual stand of these living baptismal fonts, located in shallow water in the lagoon, about 16,000 feet east of the northwest end. Along the muddy bottom was a rusting pipe, its surface as though whitewashed, This encrustation wag pale green with S. cresswellii. The species (Deg. & Deg. 24,910, Feb. 1958) likewise impregnate a cuttle-bone lying dry on the beach, Schizothrix heufleri Grun. (Deg. & Deg. 24,580, Feb. aay, 1958), with Scytonema hofmannii, grew in a drying, rain-filled depression one Mile east of the north Landing Strip. It formed a thin mat over the sandy, cocoa-colored loam. Regarding another collection (Deg. & Deg. 24,600a, Feb. 10, 1958), see Nostoc calcicola. Schizothrix lamyi Gom. (Deg. & Deg. 24,598, Feb. 11, 1958) grew one mile off the north Landing Strip with Por rphyrosiphon *fuscus and Entophysalis deusta in a drying depression once filled with rain water. , ——. uve EE -22- The tough, 1-2 mm, thick, greenish film was exfoliating due to shrinkage induced by drying, and curling upward and inward to expose the cocos- colored underside and the now bare, similarly colored, exposed ground, Schizothrix longiarticulata Gardn, (Deg. & Deg. 24,567, Jan. 31, 1958), with some Scytonema hofmannii, grew inland, namely near the ancient guano railroad wharf. It formed a firm covering over the brownish sun- baked loamy sand, but was beginning to crack here and there, and ex- foliate, The same species (Deg. & Deg. 24,585a, Feb. 11, 1958), found with Calothrix crustacea, was about three miles east of the Landing Strip. Additional collections are Deg. & Deg. 24,596a, Feb. 11, 1958, with S. taylorii; and Deg. & Deg. 24,579a, Feb. 14, 1958, with Micro- coleus acutissima. | Schizothrix taylorii Dr. (Deg. & Deg. 24,596, Feb. 11, 1958), along with S. longiarticulata and Scytonema hofmannii, was collected one mile east of the north Landing Strip. These plants covered the loamy bottom of a drying depression of rain water with a thin film. This acted as a dustbinder only until thorough drying caused the film to crack, curl and exfoliate.” ‘Another collection (Deg. & Deg. 24,582, Feb. 10, 1958), formed a greenish film over the cocoa-colored mud of a shallow arm of the lagoon, on the south side of the atoll. Spirulina subsalsa Cerst. (Deg. & Deg. 24,753, Feb. 12, 1958) was brought from a depth of ten to fifteen feet for us in the lagoon by Mr, Albert Lincoln by skin diving. Soirulina - tenerrima Kutz. Care & aes 24,785, Feb. 6, 1958) grew in association with Hydrocoleum 1 yngbyaceum Kiitz., on a smooth, lime- oo ocean reef along the north shore, Sumploca hydnoides Kiitz. (Deg. & Deg. 24,750) was collected under similar circumstances as Spirulina subsalsa. NOSTOCACEAE Nodularia sphaerocarpa Born. & Flah. (Deg. & e. 2h, 599a, Feb. 11, 1958), see see discussion of Phormidium ‘papyraceun, Nostoe calcicola Breb, (Deg. & Deg. 24,600, Feb. 10, 1958), in associa- tion with S$ Schizothrix heufleri, grew near the ruins of the PAA hotel, south of Musick Light. Here we found a concrete platform, exposed to full sunlight. A few of the cement squares had tipped in settling, enabling a rain puddle to form. Here grew the above plants. SCYTONEMATACEAE Plectonema nostocorum Born. (Deg. & Deg. 24,599b, Feb. 11, 1958), see discussion ofPhormidium papyraceum. Plectonema terebrans Born. & Flah. (Deg. & Deg. 24,573c, Feb. 8, 1958), see discussion of Jo Johannesbaptistia pellucida; for Deg. & Deg. 2h, 595a, Feb. 1, 1953, see treatment of Microcoleus tenerrimus, 22ge Scytonema hofmannii Ag. (Deg. & Deg. 24,572, Feb. 2, 1958) is probably the most persistent sand-binder on Canton, and fortunately common. It is to be seen about the CAA residential area along and even on the white roads, where pincushion-like tufts of Fimbristylis pyénocephala and the more open F, dichotoma grow, Omitting the larger projecting pebbles, gleaming white in contrast, it covers the surface of the ground with ah almost black matrix, When we drew the attention of several residents to these Scytonema areas over which they had been walking day after day, they were aStounded to eae that what they had always considered a tar- rich nes 8 was adtual y a tolony of "seaweéd or limu." With the Scytonema grew séme Microcoleus chthonoplastes. These areas in February were alternately sun-scorched and rain-drenched. Additional collections of this species are Deg. & Deg. 24,567a, Jan. 31, 1958, with Schizo- thrix longiarticulata; Deg. & Deg. 24,580a, Feb. 11, 1958, with Schizo- thrix heufleri; and Deg. & Deg. 24,596b, Feb. 11, 1958, with Schizothrix taylorii. RIVULARIACEAB Calothrix crustaceae Thur. (Deg. & Deg. 24,585, Feb. 11, 1958), with Schizothrix longiarticulata and the almost ubiquitous Entophysalis deusta, was loosely holding the white, muddy sand in a drying depression three miles east of the north Landing Strip. It (Deg. & Deg. 24,58la, Feb. 15, 1958) grew with the same E. deusta in a drying fresh water puddle one mile west of Shark Pool, southeast tip of Canton. For specimen Deg. & Deg. 24,595c, Feb. 1, 1958, see discussion of Micro- Coleus tenerrimus. Lichens, mosses and ferns have not been found on Canton. tape. REFERENCES CITED Dawson, E. Y. Some marine algae from Canton Atoll. Atoll Res, Bull. 65: 1-6 , 1959. Degener, E. and Degener, I, The Hawaiian beach Scaevola (Goodeniaceae). Phytologia 6(6): 321, 1958. Degener, 0. and Fosberg, F. R, A Central Pacific Sesuvium. Occ. Pap. Bishop Mus. 21: 45-47, 1952. Degener, 0. & Gillaspy, E. Canton Island, South Pacific. Atoll Res. Bull. 41: 1-50, 1955. ) Fosberg, F. R. Pacific forms of Lepturus R, Br. (Gramineae). Occ. Pap. Bishop Mus, 21: 285-294, 1955. Fosberg, F, R, and Sachet, M.-H, The Indo-Pacific strand Scaevola. Taxon 5: 7-10, 1956. Hansen, I, and Potzal, E. Beitrdge zur Anatomie und Systematik der lepturae, Bot. Jahrb. 76: 250-270, 1954. Hatheway, W. H, The natural vegetation of Canton Island, an equatorial Pacific atoll. Atoll Res. Bull. 43: 1-9, 1955. Van @valuwenburg, R. H., The insects and certain other arthropods of Canton Island, Atoll Res, Bull. 42: 1-11, 1955. Webster, G. L. Studies of the Euphorbiaceae, Phyllanthoideae II, The American species of Phyllanthus described by Linnaeus. Jour. Arnold Arb, 37: 1-14, 1956. ATOLL RESEARCH BULLETIN No. 65 Some Marine Algae from Canton Atoll by E. Yale Dawson Issued by THE PACIFIC SCIENCE BOARD National Academy of Sciences--National Research Council Washington, D. C. May 15, 1959 Some Marine Algae from Canton ato12/ by E. Yale Dawson The following annotated list has resulted from an examination of a collection of algae made during the first part of February, 1958, by Dr. and Mrs. Otto Degener on Canton Atoll in the Phoenix group. Only the green, brown, and red algae are treated here, inasmuch as a rather large series of Cyanophyta has been identified by Dr. Francis Drouet and is ac- counted in Bulletin No. 64. With the exception of about a dozen species mentioned by Degener & Gillaspy (1955) on the basis of determinations by M. S. Doty, this is the first floristic list of which I am aware of the marine vegetation of Canton Atoll, or of any island in the Phoenix group. Accordingly, these records supplement and extend our knowledge of central Pacific atoll floras derived from such reports as have appeared in recent years for the Marshall, Gilbert, and Line Islands (cf. Taylor 1950, Dawson, Aleem & Halstead 1955, Dawson 1956, 1957, Moul 1957). The specimens are cited according to Degener collection numbers. The first set of specimens, all of which are liquid preserved, has been deposited in the Herbarium of the University of California, Berkeley. CHLOROPHYTA Enteromorpha clathrata (Roth) J. Ag. 24841. Enteromorpha sp. 24863. This appears to agree with what Chapman has called E. clathrata var. pumila (Aresch.) Chapm. from New Zealand. Enteromorpha kylinii Bliding 24763; 24764; 24739; 2ek8uo? Ulvella lens Crouan 24843a, growing on Caulerpa serrulata. Cladophora fascicularis (Mert) Kutz. 24717. This material, about 15 cm. tall, is in good agreement with Borgesen's illustrations of specimens both from Mauritius and the West Indies. The specimens are much smaller in height than many West Indian ones, but agree in cell size and form. Cladophora flexuosa (Griffiths) Harv. 24852. The specimens show such close accord in size, habit, branching and cell form with this tropical Atlantic species as to be referable here with reasonable certainty. The similarity to Vicker's illustration of the species from Barbados is especially striking. The plant is reported from Australia, but apparently not from the central Pacific. 1. Contribution from the Beaudette Foundation for Biological Research, Solvang, California. be Dictyosphaeria cavernosa (Forrsk.) Borg. 24724; 24735; 24832; 24839; 2l865. Boodlea composita (Harv.) Brand 24716b; 24719. ‘The latter speci- men represents a lax form and bears much sterile Ceramium gracillimum v. byssoideunm. Cladophoropsis gracillimum Dawson 24758. These are in good agree- ment in structure, but are slightly coarser than either the Mexican type or material from Arno Atoll, Marshall Islands, and much coarser than Eniwetok material. The colony is quite com- pact and the filaments densely arranged in a rather erect fashion. The exceedingly long cells and thick, stratified walls are dis- tinctive. Number 24781 is still coarser, up to 140 p in diam- eter in some places, although mostly about 125 p. Some of the walls are up to 20 u thick. This would seem to be a particu- larly robust form of this generally slender species. Cladophoropsis sundanensis Reinbold 24715. This is very much like material from Palmyra Island, forming rounded, compact colonies as reported from there. The filaments are about the same size or tending to be slightly larger. 2482Ga, growing in a turf with Jania tenella, has filaments 70-100 u in diameter. 2h747 is a very young colony compacted with Lyngbya filaments. 24766 shows the characteristic sub-spherical clumps, but the cells are very long. 24820 has filaments 100-130 p in diameter. 24846 is a darker colored form with somewhat shetes cells than other ma- terial from this area. It may be a distinct entity, but points of distinction are not clear. Derbesia attenuata Dawson 2572a. Well developed, typical ma- terial, on furbinaria. Caulerpa racemosa var. peltata (Lamx.) Eubank 24713; 24716; 2h756a; 2787; 24831; 24845; 24861; 2k86h. Caulerpa racemosa var. turbinata (J. Ag.) Eubank 24784. Caulerpa serrulata (Forrsk.) J. Ag. emend. Bérg. 24751; 24756; 2h782; 248k3. ; i Caulerpa urvilleana Mont. var. 24840; 248he. Bryopsis pennata Lamx. 24731; 24743 (On Turbinaria); 24822a; 243837; 24850. Unilateral branching is especially prominent and consistent in this last collection. . The following identifications and notes on Codium are provided by Dr. Paul C. Silva, University of Illinois. Codium ovale Zanard. 2746; 2h7hO. This latter collection is a mixture of two species growing in intimate association: Codium ovale and a member of the C. arabicum complex. The known range of C. ovale has now been extended beyond the type locality -3- (New Guinea) to include Kwajalein, Majuro and Canton islands. The adherent Codium is somewhat more cerebriform than is usual for C. arabicum, but anatomically it clearly reveals its relationship to the C. arabicum complex. Codium geppii 0. C. Schmidt 2714; 24570; 24788; 24821; 24851; 24825. These are very similar to the type collection. In the Indo-Pacific region a member of the C. arabicum complex and one of the C. geppii complex are always found as an integral part of the biocoenosis of coral reefs. In the Atlantic a similar, but less constant, relationship holds: an adherent form, referable to the C. intertextum complex, is invariably present on coral reefs, while a repent form, referable to the C. repens complex, is found only on certain islands. Halimeda fragilis Taylor 24774. Halimeda micronesica Yamada 24768. A good specimen 6-7 cm. tall. Halimeda opuntia (L.) Lamx. 24771. PHAEOPHYTA Ectocarpus indicus Sonder 24836. Good material on Turbinaria with very long plurilocular sporangia often to 200 u or more. Sphacelaria furcigera Ktitz. 24754 (on old Turbinaria); 24752b. Sphacelaria sp. 24723a. This may be a tropical form of S. sub- fusca Setch. & Gard. The specimens have propagulae like S. furcigera, but many of them are trifurcate rather than bi- furcate. Plurilocular sporangia, however, are dominant on this material. Dictyota friabilis Setchell 24845a; 24828; 24859; 24862; 24868. This latter collection is almost without doubt like the type from Tahiti. Dictyopteris repens (Okam) Borg. 24752 (small amount). Pocockiella papenfussii Taylor 282k. Pocockiella variegata (Lamx.) Papenf. 2765. Turbinaria ornata (Turn.) J. Ag. 24783. Turbinaria trialata (J. Ag.) Kitz. 2721. RHODOPHYTA Gelidium pusillum (Stackh.) Le Jolis 24736. ‘Typical, small, cystocarpic material 3-4 mm. tall. en Pterocladia sp. 24/42; 24755. These tetrasporic and cystocarpic materials are well developed, over 2 cm. tall, and pinnately branched much as in Pterocladia nana Okam. from southern Japan and Formosa. They may be referred provisionally to that species, but additional Pterocladia collections from the cen- tral Pacific, such as at Palmyra Island where one is reported from intestinal contents of fishes (Dawson, Aleem and Halstead, 1955) are needed to enable us to interpret more satisfactorily the occurrence of this genus. Gelidiella rigidiuscule (Grunow) J. Feldmann 24737. This richly fertile material has the stichidia and branching much as in G. acerosa, but is smaller in size: 300 p down to 120 p in the ultimate branchlets. This is in agreement with the discussion and key in Feldmann (1931) of G. rigidiusculum from Ceylon, ' for which reproduction is not reported. The size and habit are much like G. hancockii from the Gulf of California, but that species has cylindrical tetrasporangial stichidia on short lateral branchlets rather than acute, bulbous, terminal ones as in the present material. This disposition seems justified awaiting the report of fertile material from Ceylon for com- parison. 24838 (with some Herposiphonia secunda); 24759 tetrasporangial); 24832a (some growing with Dictyosphaeria); 24789. The tetrasporangial stichidia in this collection are so abundant as to terminate almost every branch. Gelidiopsis intricata (C. Ag.) Vickers 24728; 24857a. Jania capillacea Harv. 2857. Jania micrarthrodia Lamx. 2h761a (cystocarpic); 24737a (with Gelidiella rigidiuscula); 24860. Jania tenella Ktitz. 24826 (in mixture); 24822 (a questionable form with very abundant conceptacles). Hypnea esperi Bory? 24730; 24854. These specimens have the tet- rasporangial sori in part at first unilateral on ‘the branches, causing some distortion. Later they may fill in all around. They are similar in this character to specimens reported and illustrated by Setchell as H. nidifica J. Ag. from Tahiti. Gracilaria sp. 24866. This may be the plant described as Corallopsis reptans Weber van Bosse, from the Kei Islands. Lomentaria sp. 2h86hb. Ceramium clarionensis Setch. & Gard. 2h730b. Ceramium equisetoides Dawson 24864a. These are in good agree- ment with Pacific Mexican specimens. Ceramium gracillimum var. byssoideum (Harv.) G. Mazoyer 2h716a; 2h784a (with Caulerpa racemosa v. turbinata); 247 30 i a (a ver lax form); 2804; LBL. Sitbinate); 25730e-(e very aoe Ceramium vagabunde Dawson 24843c. This material is somewhat more slender than Eniwetok specimens and has nodal bands more like those of the material reported from Isla San Bene- dicto, Mexico, yet the agreement is eecmares satisfactory. Growing with Caulerpa serrulata. Ceramium sp. 248h0e. Fertile, tetrasporangial material appar- ently near C. personatum Setch. & Gard. The agreement is good except that the descending appendages within the axial cells were not observed. The tetrasporangia are at first abaxial, then whorled and involucrate. Growing with Caulerpa urvil- leana. Griffithsia sp. 2h486lc, sterile. Herposiphonia secunda (Ag. ) Ambronn 24733 (richly developed); 24805; 24829 (richly developed); 24834. Heterosiphonia wurdemannii var. laxa Borg. 24748 (mixed with Herposiphonia); 4741 (in mixture); 24854a (on old Turbinaria); 24853. The following determinations and notes on Polysiphonia are provided by Dr. George J. Hollenberg of the University of Redlands, California Polysiphonia ferulacea Suhr 24866a. This determination is probably correct if Tseng, Cribb and others have been correct in their identifications of Pacific specimens. It should be considered tentative, however, until male plants are found. Polysiphonia flaccidissima Hollenberg 24867. This determination is probably reliable although the segments are shorter (1.0- 2 diameters ) and the trichoblasts « are not well developed and with tapering tips. Polysiphonia mollis Hook. & Harv. ?? 24823. This identification is doubtful because of (1) small size; (2) obvious sympodial branching; (3) very long, continuous tetrasporangial branches. Laurencia nana Howe eh761. The cortical cells are of the non- palisade type, although an error occurs in expressing this in Dawson 1957, p. let. 26% LITERATURE CITED Dawson, E. Y. 1956. Some marine. algae of the southern Marshall Islands. Pac. Sei. 10(1): 25-66, 66 figs. 1957. An annotated list of marine algae from Eniwetok Atoll, Marshall Islands. Pac. Sci. 11(1): 92-132, 31 figs. Dawson, E. Y., A. A. Aleem and B. W. Halstead ~ 1955. Marine algae from -Palmyra Island with special reference to the feeding habits and toxicology of reef fishes. Allan Hancock Found. Publ., Occas. Papers (17): 1-39, 1 map, 13 figs. Degener, 0. and BE. Gillaspy ih 1955. Canton Island, South Pacific. Atoll Res. Bull. 41: 39-h2. Feldmann, J. 1931. Remarques sur les genres Gelidium Lamour., Gelidiopsis Schmitz et Echinocaulon (Kittz.) emend. Recueil de Trav. Cryptogam. dédiés a Louis Mangin. pp. 151-166. Paris. Moul, E. fT. . 1957. Preliminary report on the flora of Onotoa Atoll, Gilbert Islands. Atoll Res. Bull. 57: 1-48, Taylor, W. R. 1950. Plants of Bikini.... xv 227 pp-, 79 pls. Univ. Michigan Press, Ann Arbor. ATOLL RESEARCH BULLETIN No. 66 Notes on the geography and natural history of Wake Island by he H, Sayan, om. Issued by THE PACIFIC SCIENCE BOARD National Academy of Sciences--National Research Council Washington, D. C. May 15, 1959 Notes on the geography and natural history of Wake Island by E. H, Bryan, ir V Wake is a V-shaped atoll in the northwestern Pacific, north of the Marshall Islands, between Midway and Guam, Its direction and distance are: Direction Nautical Miles StatuteMiles Wake to Midway N, 60°R. 1029 1185 Wake to Honolulu N. 85°R. 2004, 2308 Waterto Guam N. 75°W. 1309 1508 Wake to Tokyo N. 55°W. 1723 1985 Wake Atoll consists of three islets, Wake islet, the largest, on the southeast, has the shape of a V, the arms of which are about two and three quarters land miles long. Each arm is continued, beyond a narrow lagoon entrance, as a separate islet, Peale islet on the north and Wilkes islet on the south, The western ends of these two islets are connected by a sweep of flat reef, which continues as a narrow border along the ocean Side of all three islets. In 1923, the enclosed rectangular lagoon had depths up to fifteen feet. The following is a summary of the land and salt water areas of the atoll: Area: Square land miles Acres Wilkes (named for Charles Wilkes, USN) 0.31 197.44 Wake (named for William Wake ) 2ely 1,367.04 Peale (named for Titian Peale, artist) 0.40 _ 256.83 Total land area 2neo Uy feyeilee a. Area of enclosed lagoon (water and sand flat) 3.75 Square land miles The names of the two smaller islands were given by Dr. Alexander Wetmore and other members of the U.S.S. Tanager Expedition on July 27, 1923, Lieutenant (later Commodore) Charles Wilkes was the leader of .the United States Exploring Expedition, which visited and mapped the atoll, December 20, 1841; Titian Peale was the artist and one of the naturalists on this same expedition. 1/ Curator of Collections, B. P. Bishop Museum, Honolulu 17, Hawaii. - This compilation from various notes in the author's geographic files was prepared in August 1957, so that the information could be made available to persons living on or visiting Wake. Portions are taken direct from the author's book, "American Polynesia and the Hawaiian Chain, Honolulu 1942, chapter 57, pages 208 to 214. o2en This atoll also has been known as Halcyon or Helsion, and some authorities think it may have been the same as San Francisco Island, discovered by the Spanish explorer, Mendana, October 4, 1568. The official discovery, however, is credited to Captain William Wake, in the British schooner Prince William Henry, in 1796. The island was seen in 1823 by Captain Gardner from the whale ship Bellona. , He described it as being 20 to 25 miles long /quite an exaggeration/, with a reef extending two miles from the east end, with detached rocks on the west [probably those on the curving reef/, He noted that it appeared well covered with trees, It was also seen by Cuptain James Hunnewell from the Mentor, December 20, 1824. Halcyon Island was said by Captain Kotzebue to have been an American discovery, located at about 19° 23! N., 1659 33' E. After unsuccessful search for it'by Captain Sproule of the barque Maria, by Cuptain Brown in the Morning Star, and by the U.S. Exploring Expedition, the conclusion was ‘reached that Halcyon was the Same as Wake, Captain F. W. Beechey, R.N., in H.B.M. ship Blossom, tried to locate Wake Island in March 1827, without success. The United States Exploring Expedition, under Charles Wilkes, visited and mapped Wake in December 1841. Wilkes! "Narrative" (V: 284- 5, 1844) reads:- i "Wake's Island next claimed my attention. On the 19th we reached. its parallel, and hove-to till daylight of the 20th, when we discovered it, bearing west-by-north, about nine miles distant. The wind was light from the north-northeast. After breakfast, several boats were sent to survey the island. Wake's Island is a low coral one, of triangular form, and eight feet above the surface. It has a large lagoon in the center, which was well filled with fish of a varjety of species; among these were some fine mullet. There is no fresh water on the island, and neither pandanus nor cocoa-nut trees, It has upon it the shrubs which are usually found on.the low islands of the Pacific, the most abundant of which was the Tournefortia. Mr. Peale found here the. Short-tailed albatross, and procured an.egg from its nest.- The birds were quite tame, although they were not so numerous as we had before met with on uninhabited islands. "The time of low water took place at one o'clock, and the moon entered its last quarter on the same day: the tide was setting along the shore of the island with much strength to the westward; the rise and fall was three feet, From appearances, the island must be at times submerged, or the sea_makes a complete breach over it; the appearance of the coral blocks and of all the vegetation leads to this conclusion, for they have a very decided inclination to the eastward, showing also that the violent winds or rush of the water, when the island is covered, are from the westward. The reef around this is- land is very small in extent. "The position of Wake's Island was found by my observations of equal altitudes on shore to be in longitude 166° 31' 30" E., and latitude 199 10' 54" N. "By four o'clock, P.M., all the boats had returned on board, when we filled avay and proceeded on our course to the westward, Although these coral islands resemble one another very strongly, yet they afforded us some recreation for a few hours, and much satisfaction in obtaining get series of observations in magnetism. Our visit to Wake's Island gave us an opportunity of adding to our collections in natural history. "In the evening we steered to pass over the position of Halcyon Island,- longitude 163° 30! E., latitude 19° 13' N.; and on the 27th, we passed immediately over its locality, and had run on its supposed eereniet fifty miles on each side of it, but nothing was seen of it. We now felt the current to the southeast 12' in the twenty-four hours." Following the careful mapping of the island by Wilkes in 1841, several vessels are recorded as having sighted Wake, including the barque Maria, under Captain Sproule, in 1858; as noted by Dr. William T, Brigham, Wake was seen also "from the masthead of the ship Oracle in 1865." On March 4, 1866, the Bremen barque Libelle, under command of Captain Tobias, went ashore on the east reef. On board were several prominent passengers and a cargo valued at over $300,000. Among the - passengers were Madame Anna Bishop, Miss Phelan, M. Schultz and Charles Lascelles, of an English opera troupe, a Japanese traveller named Kisaboro, and Eugene M,_Van Reed, whose account of the experience appears in the Friend /fonolul yy for September, 1866, Following a hazardous night on on the ship, during which waves broke over the vessel, passengers and crew were landed the following day with great diffi- culty through the breakers, After three weeks on the island, without finding source of food or water, it was decided to try to reach the Mariana Islands in open boats. On March 27 they set out, passengers in the 22-foot longboat, twenty-two persons under command of the First Mate, and the Captain and remainder of the crew in the gig, with what provisions and water they were able to salvage, After thirteen days of frequent squalls, short rations, and tropical sun, the longboat reached Guam. The Captain with eight persons, in the twenty foot gig, were not heard of again, although a schooner from Guam went in search. The passengers were strong in their praise of the courtesies received from Francisco Moscoso y Lara, Governor of the Marianas. Several veSsels went to Wake to salvage the cargo, which included several hundred flasks of quicksilver, The sloop Hokulele, with a party headed by T. R. Foster, left Honolulu May 9, 1867, “reached Mieke on May 3lst, left there June 22, and returned to Honolulu July 29, with 247 flasks of quicksilver, A brig from China salvaged another 248 flasks at about the same time. Thomas Foster, Captain English, and eight Hawaiian divers landed at Wake from the Hawaiian schooner Moi Wahine in Septem- ber 1867. Three days after their arrival their schooner, with Captain Zenas Bent in command, mate Wight, and a crew of five, was driven to sea by a gale and not heard of again. The salvage party was rescued by the English brig, Cleo, Captain Cargell, in March 1868, and returned to Honolulu on April 29, with 240 flasks of quicksilver, some copper, anchor and chain. In 1883 the German warship Leipzig passed close to Wake and a careful determination of its position was made. During the Spanish-American war, several vessels going to and returning from the Philippines stopped and raised the American flag. One of these visits, perhaps the earliest, was on July 4, 1898, by ae General F, V. Green, commanding the second detachment of the Philippine expedition, from the S, S. China. Another, also in July 189%, was by General Merritt, from the U. S. Army Transport Thomas. This landing probably was made in the little cove near the eastern end of Wilkes islet, for on August 2, 1923, just inland from the landing place, the writer found a section of flagpole, about'18 or 20 feet long, on which was burned in block letters, "U.S.A.T. Thomas," The formal annexation of the island by the United States took place. on January 17, 1899, according to an account by Commander Edward D, Taussig in the U.S. Naval Institute Proceedings for June 1935. He commanded the U.S.S. Bennington which made the voyage from Honolulu to Wake for that purpose. The landing was made in the cove noted above, and at 3:22 P.M. the American flag was hoisted by Ensign Wettengell and a salute of 21 gums fired from the Bennington. The position of the flagstaff was determined, from observations on the ship, to be: 199 7! 50" North, 166° 31' Fest. The account continues: "After the salute was fired the flag was nailed to the masthead with batten, and a brass plate with the following inscription was screwed near the base of the flagstaff: United States of America William McKinley, President; John D. Long, Secretary of the Navy, Commander Edward D, Taussig, U.S.N., Commanding U.S.S. Bennington, this 17th day of January, 1899, took possession of the Atoll known as Wake Island for the United States of America." During the next decade an occasional American ship stopped, but there is very little recorded history. Vuring this time the island was visited by Jepanese poachers, collecting the feathers of sea birds for millinery purposes. Two camps were established: one on the eastern end of Wilkes islet, where the Tanager Expedition in 1923 found a single wooden shack and a grave; and one across the lagoon near the eastern end of Peale islet, where there was a more extensive camp. Tais is described in the writer's field notes for July 31, 1923, as follows: "The camp consists of the remains of two large frame buildings with galvanized iron roofs, about 18 feet wide, one 20 feet long, one 30 feet long; two smaller buildings; one tank, and one storehouse, raised on posts which are guarded with tin. Scattered about were a number of barrels, boxes, two large clay water jars, tin cans and metal kettles. Suw part of a Sydney newspaper, a pile of oakum, bamboo frame with lath trays, There was also a boat, a little larger than a skiff. Made a copy of a Japanese inscription inside the bunk house." Later this was translated to read something about leaving the island, with the date, November 13, 1908. In 1912, the U.S.S. Supply stopped at Wake Island. A whaleboat landed some men who planted coconut palms brought there from Guam. No Sign of these was seen in 1923, "1 bi mas The Tanaser Expedition made an extensive biological survey of Wake from July 27 to August 5, 1923. ‘!neir camp was along the ocean beach opposite the landing place at the eastern end of Wilkes islet. A map of the atoll was made by James B. Mann and Professor Harold S, Palmer, to which the writer added determinations of latitude and longitude, made from a boulder near the camp. Meanwhile soundings were made around the island from the U.S.S. Tanager, under command of Lt. Cmdr. Samuel Wilder King, later Delegate to Congress and Governor of Hawaii. Although the vessel worked as near the reef as it dared, at only one spot was it possible to reach bottom with 100 fathoms of line; this was about 1500 feet off Heel Point, where a sounding of 85 fathoms was made. A total of 19 species of flowering plants was found growing naturally on Wake. Much of the surface of all three islets was found covered by scrub forest, 12 to 20 feet high. Same of the forest was so dense that one could not walk through it with speed or comfort. Other places, such as the middle portion of the northeastern arm of Wake and the western ends of Wilkes and Peale, there were areas where the trees were low and scattered, with rocky surface and scrubby undergrowth, as if here the sea broke across the rim at time of storms. eve ees eee heliotrope, also known to scientists as Messerschmidia argentea, a species widespread on Pacific Islands. It grows to a height of about 20 feet, with an umbrella-shaped canopy of rosettes of large leaves, covered with silvery hairs. Even larger in size, but confined to the northwestern end of Wake islet was the “buka‘ tree (so-called by Gilbert islanders), Pisonia grandis, with massive trunks of very soft wood and sticky flowers and fruit. On Wilkes islet, and apparently spreading rapidly along the lagoon beach of Wake, to the east and north, were tall wiry bushes of Pemphis acidula. In the interior of Wake were small clumps of kou trees (so-called in Hawaii), Cordia subcordata, a hard wood tree, much prized in some regions for woodwork, but here scrubby and worth- less. Growing over trees, rocks and bushes, and forming tangles on the ground, was a morning-glory vine, Ipomoea tuba, formerly called Ipomoea grandiflora. The plants observed in 1923 have been listed by Christophersen (1931). The bird life on Wake consisted of about a dozen species of sea birds, half a dozen migratory species, and the flightless rail, Rallus wakensis, the only native land bird. The only mammal was the Polynesian rat, Rattus exulans. Of reptiles there were two kinds of geckos and two kinds of skinks, to be described later. A number of species of insects were collected and many kinds of marine life, both fishes and inverte- brates, To those of us who camped on Wake in 1923, as much a pest as the rats were the hermit crabs. One large, red-legged species, Dradanus unctulatus, got into our provisions, ran off with our soap, and combined the activities of pack rats with those of being the garbage department of the island. In 1935 Wake Island was placed under jurisdiction of the United States Navy Department by Executive Order, That same year Pan American =—On Airways established a modern airport, using the south shore of Peale islet, just west of Flipper Point, as the site of pier, shops, water tanks, and a modern hotel. Then the threat of war descended on Wake. On February 14, 1941, President Franklin D. Roosevelt signed an Executive Order making the island a national defense area, The military importance of Wake Island led to great activity on the part of the United States to fortify the island during 1940 and 1941. The defense force was small and although gallant, not equipped to defend the island against overwhelmingly superior numbers. It is not an object of this account of the natural history of Wake to expand upon the World War II's history of the island. That has been done by Major James P, S, Devereux, the commanding officer of the Marine forces (1947), by R. D. Heinl, Jr., in the official Marine account of the defense of Wake (1947), by Rodney Kaphart (1950), and D. C. Woodbury (1946). There is also a brief official Naval release in the Paradise of the Pacific for February 1942 /pages 10 to 12/. The air attack began on December 8, 1941 and continued for several days as wave after wave of Japanese planes bombed the little island. The end is told in a historical marker, set up following the war on the south shore of Wake islet:- "In this area occurred the principal fighting which led to the island's surrender. About 1000 Japs were landed between 0120 and 0430, 23 Dec. 1941. They were opposed by 85 U.S. Marines who fought ferociously and valiantly until 0030 in the best traditions of their corps until order- ed by their superiors to surrender to save American life because the situation was hope- less, Marine forces were composed of portions of the First Marine Defense Battalion under Major Devereux and Marine Fighting Squadron 211, under Major Putnam." The Japanese renamed Wake "Otori" and held the island until the end of the war, surrendering to an American force September 7, 1945. The capture achieved very little for the Japanese. The island was bombed repeatedly by American forces and its isolation made it diffi- cult to supply. The garrison, at times numbering more than 4,000 ; Japanese, suffered great privation from lack of food and water. Mute evidence of the Japanese efforts to fortify Wake, and the American work of destruction is to be seen everywhere on the island. Pathetic little garden plots tell their own story of starvation. Wrecks of Japanese vessels dot the beaches. One of the blackest and saddest paragraphs in the whole war story has to do with the fate of the 98 American civilians, who could not be evacuated by the Japanese. They were marched to the beach, and es.) ine en LG, there cruelly slaughtered by command of Admiral Sekaibara, who after the war was tried and hanged for this act. Wake Island came into the limelight in October 1950, when it was the meeting place for President Truman and General Douglas MacArthur, in their discussion of the Korean situation. Another event, which will long be remembered by the 550 persons who were on Wake, was typhoon Olive which hit the island in September 1952, It cost some $1,600,000 to rebuild installations destroyed or damaged by this tropical storm, | At present (1957) Wake Island is in the charge ofi the Civil Aeronautics Administration, which maintains and controls the airport and runway. Four short visits were made to Wake by F. R. Fosberg during which he made collections, principally of plants, and observations am the vegetation and on the general ecology of the atoll. These were, in order, on October 23, 1951, April 19-21, 1952, July 18, 1952 and October 22-23, 1953. His observations on birds and other vertebrate animals, with a section on pest control, have been incorporated into the present article, For his account of the vegetation and flora see the separate article by him immediately following this one. Notes on rats and pest control on Wake Island, 1952.--Contributed by F. R. Fosberg On a brief visit in October, 1951, rats were noticed in the Pan American Airways dining hall, but no particular attention was paid to them. They were at that time or somewhat earlier, said to be so numerous that they bothered the Pan American plane crewmen sleeping in the hotel. In April, 1952, two days were spent on Wake, much of the time in the field with Fred Schultz, at that time in charge of pest control for the Civil Aeronautics Administration on the island, During the previous year Mr. Schultz had waged an extremely successful campaign against rats on the island. He started by studying the habits of the rats rather carefully. He found that they lived largely in the clumps of bushes, especially those which were covered by tangles of the wild white morning glory, Ipomoea tuba. He found that they utilized the enlarged immature fruiting calyces of this species in lieu of a water supply. He said there were four kinds of rats on the island, In the two days of actively looking for rats, along with other animals and plants, only one rat was seen, this a large blackish one. Schultz had for many months carried out an active poisoning program, uSing Warfarin in rolled oats, placed in small quonset—hut- like shelters about a foot long, and renewed frequently, These Mr. Schultz strung in two series of lines across the center of the island from lagoon to outer shore a short distance apart. These were visited =-5— every day, examined, the bait renewed, till the baits were not dis- turbed much any more, Then the series of lines were moved a little farther apart. This was continued until the entire atoll had been covered, and the rats greatly reduced. After this a few lines were maintained, scattered at intervals over the islands, and visited and the bait renewed once a week, As long as less than 10% of the baits were disturbed the rats were considered controlled. Flies and mosquitoes were also the special objects of Mr. Schlutz! efforts, At the beginning of 1951 there were untold numbers of these insects on Wake Island. Garbage and other fly breeding material was eliminated. Garbage dumps were regularly burned and buried. Old tires, cans, and other breeding places for mosquitoes were systematically drained and punctured, and larger bodies of water, such as ponds, bomb craters, and cisterns were stocked with mosquito fish. Three species of mosquitoes had been known from the island, Culex quinquefasciatus, Aedes aegypti, and another Aedes, possibly A. scutellaris or A. albopunctata, The success of these measures may be judged by the fact that in two and a half days spent in covering the island, mostly. in company with Mr. Schultz, only one rat, no mosquitoes, and a very few flies were seen. In his weekly examinations of his poisoned baits at this time he commonly found less than 5% of them touched by rats. LAND _AND_FRESHWATER VERTEBRATES MAMMALS In 1952 domestic dogs and cats were seen by F. R. Fosberg, but none were specifically noted in 1953. Although Mr. Fred Schultz stated that prior to his activities, described abové, there were four species of rats on lake, a check by Dr. D. H. Johnson of the collections at the U. S. National Museum shows that only Rattus exulans Peale is represented by specimens from this island. He suggests that the others noted by Mr. Schultz were probably forms of Rattus rattus, the common house rat, though it is possible that Rattus norvegicus, the Norway rat, could have established itself there also, A series of skins and skulls collected to represent the rat populations living on the‘island would be very desirable to settle this question. BIRDS Now that the Wake Island rail is gone, only sea and migratory birds are found on Wake Island. Dr. Alexander Wetmore made a careful survey of the birds of Wake during July and Aueust 1923, but no. detailed report resulted. The rail was present at that time. In 1940, the writer prepared a list of the birds likely to be seen on Wake for Torrey Lyons, who was working on the island, but this was made from memory and notes, without an actual survey. In 1952, Dixon and ie, i | H 29. Starrett reported various birds seen at sea near Wake during 1945 and 1946, Writing in the Elepaio for January 1953, H. Paul Porter praised "the brave young birds of Wake Island," on their ability to ride out a typhoon which left some ten million "dollars damage in its wake; but he mentioned only "terns." The only detailed recent account of the birds which I have seen gives a rather discouraging picture of what the war left on the island. | Dr, Alfred M, Bailey, "Notes on the birds of Midway and Wake Islands," | 1951, says: that the main island is nearly devoid of bird life, other | than a few sea and migratory species. Wake rails had not been seen | Since the island was reoccupied by Americans. He quotes from the diary of a Japanese officer regarding daily bombing by American planes. "An order has just come out," states the diary, "forbidding us to catch gooney birds Weibatnedsbey lest they be wiped out." Dr, Bailey continues: "The effort of this officer to protect the 'gooney birds' must have been largely in vain. Most of the birds were destroyed by the starving soldiers, although a great colony of sooty terns (Sterna fuscata) was guarded so that the eggs could be gathered regularly. The Booty tern colony on Peale Island ... was the largest I had ever seen. On May 15 L194) I saw thousands of birds on their eggs. The downy young were beginning to hatch on that date, on the rusted iron skeletons of b eres naval raeeTataee, Small groups of noddy terns (Anous stolidus) were nesting in the dwarfed trees known locally as bukas (Pisonia grandis). I saw two white-tailed tropic- birds, It was apparent that the bird population had suffered from the Japanese occupation, for aside from what I have just mentioned, there were no birds, We did not see any boobies (Sula), red-tailed tropic- birds, or albatrosses. A search for such burrow-nesting birds as petrels and shearwaters was out of the question, for my time was so short. I saw nothing of either." Despite this discouraging statement, I venture to repeat my list of birds which might be expected to occur on Wake, with brief descrip- tion of each:- Family DIOMEDEIDAE, albatrosses Diomedea immutabilis Rothschild, the Laysan Island albatross, and Diomedea nigripes es Audubon, the black-footed albatross, might possibly stray as far as Wake from their regular area between the N. W. Hawaiian and Bonin Islands. Family PROCELLARIIDAE, petrels and shearwaters Puffinus pacificus chlororhynchus Lesson, the wedge-tailed shearwater, is the most likely member of this group; 17-19 inches long, it is a large shearwater, with long, wedge-shaped tail; upper parts sooty brown, crown neck and wings darker, forehead paler; under parts paler than upper, some bellies quite white. WS en ee ree mtr tener ene ee ee ee * One can be seen in front of hotel in post-war photo (Kaucher, 1947). DA GUA ees eee Manes ene he br inl we ee Eos Pterodroma hypoleuca hypoleuca Salvin, stout-billed gadfly petrel or onin Island petrel; 12-14 inches long; upper varts grayis:., deen slate on forehead, margined with wae under parts white except sides of breast sooty black, Family PHAETONIDAE, tropic birds Pheethon rubricauda rothschildi (Mathews), red-tailed tropic bird; ~~ 18 inches long; including long, slender tail feather, black shaft, bright red webs; plumage silky white, some tinged with rose; a few. black spots, especially in young. Nest a hollow in sand, secluded beneath a bush, bunchgrass, or slab of rock, Family SULIDAE, boobies and gannets Sula sula rubripes Gould, red-footed booby; 23-28 inches long; adult to. head and neck tinged buffy, wings with hoary-gray markings; tail white; feet red. Young: sooty-—brown above; head, back and lower parts smoky-gray; recognized by red feet, Nests in any available bush or tree, almost never on ground; lays a single egg in a crude platform of leaves and twigs. One seen returning to the island just before sunset, on July 18, 1953, by Fosberg. Sula leucogaster plotus (Forster), brown or brown-vested booby; 30-31 inches long. Adult head, neck, back, wings and upper breast deep sooty brown; all white below. Young: brownish- gray, upper parts darker than lower. Builds nest of twigs on ground or low scrub; lays two bluish-white, chalky eggs. In . 1953 Fesberg saw a number of brown boobies flying on Peale Islet. In Avril 1952 boobies were seen on Peale Islet, nesting in trees, but the species was not determined with certainty. Sula dactylatra personata Gould, masked or blue-faced booby; 25-29 inches long. Adult: white, with wing-coverts, primaries, secondaries and most of tail sooty-—brown; mask bluish or olive, legs and feet deep brown. Young: upper parts grayish-brown, white below, Lays two limy-white eggs on bare sand. In 1953 Fosberg saw a very few blue-faced boobies on Peale Islet. Family FREGATIDAE, frigate or man-o'-war birds Fregata minor (Gmelin), Pacific man-o'-war bird. Adult male: large, blackish, with forked tail; head and back glossy purple; breast lighter than belly; large red patch under chin which inflates like a child's balloon. Female like male but head blacker, chin and throat grayer, breast whiter, Exact subspecies is uncertain; could be minor of western Pacific and Indian Ocean, or palmerstoni of Hawaii, Nestling is fluffy, white "puff-ball;" becomes rufous or cinnamon before becoming black adult. Both parents sit on nest, which is large pile of sticks. Obtains fish by harassing other birds, especially boobies, in air. In 1952 many frigate birds were seen by Fosberg, mostly roosting on the remains of an old pier on Peale Islet. In 1953 there were many more, hundreds of them, flying or roosting on an old steel frame or on power lines. Salinas Family RALLIDAE, rails, gallinules, coots Rallus wakensis (Rothschild), the Wake Island rail; 9 inches long; upper parts dark ashy brown, chin and upper throat whitish, neck gray, under parts ashy brown, barred with white. Wings about 4, inches long and_so soft as to_suggest little power of flight. Endemic to Wake /See Mayr, 1945/, Believed to be extinct. Family CHARADRIDAE, plover Pluvialis dominica fulva (Gmelin), Pacific golden plover; 10 inches long; summer adult: mottled black above with golden and ashy; dull dusky black beneath; whitish frontal band and eyebrow. Winter adult (plumage most likely to be seen): without black on under parts, whitish on throat and belly; light brown, streaked with gray, elsewhere. No hind toe. Arrives from north in fall; flies north in spring. Breeds in northern Asia and Alaska. On April 20, 1952, only two plovers were seen by Fosberg, but he was informed by Mr, Fred Schultz that there were many of them only a day earlier, In October, 1953, several dozen plovers were seen feeding in ponds, and a few scattered elsewhere. Family SCOLOPACIDAE, snipe, sandpiper Numenius tahitiensis (Gmelin), bristle-thighed curlew. Length 17 inches; bill 3- inches. Dusky brown above, varied with buff; tail ochraceous, barred dark brown; dull buff beneath; cheeks, neck, chest with brown markings; thighs with bristle-like points. Migratory species. Heteroscelus incanus (Gmelin), wandering tattler; 10-12 inches long; olive-gray above, white below and on throat and chin; lower parts streaked and barred with dusky in summer (when arrive from north also), but soon molt and lose this. Feeds close to the beaches on small warine creatures. One tattler was seen by Fosberg in April, 1952, and one in October, 1953. Arenaria interpres (Linnaeus), turnstone, $ inches long; in winter: dusky brown, feathers edged with ashy-brown; abdomen white; upper parts marked with black and rufous or chestnut. In summer: sides of neck and breast black. Migratory species, usually seen in small flocks, which rise together. On wing recognized by distinct flicker of black, white and (in summer) chestnut. Frequents sea beaches and mud flats, turning stones in search of worms and marine creatures; roosts at night on rocky points near shore. One turnstone was seen by Fosberg in October, 1953. _ Crocethia alba (Pallas) /formerly called Tringa arenaria/, sanderling; 7-8 inches long; in winter: pale gray above; head, back, rump with black central area on each feather; under parts white; bill and feet black. In spring, becomes spotted and streaked with black. Catches marine organisms following retreating waves, on beaches. Migratory species. -12- Erolia acuminata (Horsfield), sharp-tailed sandpiper; 8 inches long robably rare/. Winter adult: grayish-brown above, streaked and striped dusky; white stripe over eye; chest sides and breast pale grayish-buff; abdomen white. In summer: upper parts brighter, with rusty and black; lower parts streaked dusky and grayish-brown V-shaped marks. Migratory species Siberia to New Zealand. /Other possible migratory species which might visit Wake/ Green-winged teal, Anas crecca carolinensis Gmelin Pintail /duck/, Anas acuta tzitzihoa Vieillot Widgeon /duck wea Anas penelope Linnaeus ° Canvasback /duck/, _Aythya valisineria (Wilson) Black-bellied plover, Squaterola squaterola (Linnaeus) Semipalmated plover, Charadrius hiaticula semipalmatus Bonaparte Mongolian dotterel, Charadrius mongolicus stegmannii Stresemann Whimberel, Numenius phaeopus variegatus (Scopoli) Greater yellow-legs, Tringa melanoleuca (Gmelin) Family STERNIDAE, terns Loy some included with gulls in the Laridae/ Sterna fuscata oahuensis Bloxam, sooty tern; very abundant, according to Bailey. Sooty-black above; forehead, sides of head, outer tailfeathers and underparts white; bill and feet black; 15-17 inches long. Lays one creamy-white egg, with variable spotting, in nest among bunchgrass, Catches small squid and fishes. In Avril 1952 a considerable rookery of sooty terns was observed by Fosberg on the lagoon side of Peale Islet. There were many young, varying from the pin-feather stage, light gray downy beneath and sooty brown on back and sides, to fully feathered, sooty brown speckled with white on the back, sooty gray beneath. This was not an especially large colony compared with those seen in the northern Marshall Islands. On July 18, 1952, C. J. Johnson reported that these birds had eggs. In 1953 this colony had moved to near the west end of Peale Islet after having been deliberately driven away from their old nesting place near the LORAN station. Their nesting was said to have been finished in July. In October a small number of adult birds were seen around the west end of Peale Islet. On the nesting site here were a small group of almost fully grown but more or less crippled young birds, also some dead ones, There was no obvious reason for the condition of these young birds. Sterna lunata Peale, gray-backed or spectacled tern; 16-17 inches long; forehead, broad stripe over eye and underparts white; stripe through eye, top of head and nape black; upper parts dark ashy, paler on back of neck; bill and feet black. Anous stolidus pileatus (Scopoli), noddy tern; 13-17 inches; sooty brown, grayer on neck, light gray on forehead. Young birds Similar, but lack white on crown. Eggs less speckled than those of other terns, one laid in nest on ground. In April 1952 a few common noddies were seen by Fosberg, with one nest, but this may not have been occupied. In October, 1953, many more were seen on Peale Islet, roosting or flying with the frigate birds. One nest had a half-grown young. aioe Anous tenuirostris marcusi (Bryan), white-capped noddy; 13 inches; resembles noddy, but smaller; back, neck and under parts sooty black; forehead and crown white, becoming grayer on nape to merge with blackish on shoulder; narrow black mark over eye, white streak below eye, Prefers to nest on bushes; lays one &gé- Gygis alba candida (Gmelin), white tern, fairy tern or "love bird"; pure white except for black ring around eye and black feet. Balances single egg on limb of tree or bush fork, A few fairy terns were seen by Fosberg in both 1952 and 1953, mostly ae / around the Pisonia forest. /Other terns which might visit Wake Island/ Sterna sumatrana Raffles, black-naped tern; small, forked tail; plumage white, parts with pinkish cast; back, rump, tail, Wing-coverts pale pearl-gray. sterna anaetheta Scopoli, bridled tern; like sooty tern but smaller, 14-15 inches, narrow white band on forehead extending back behind eyes; grayish brown below. Thalasseus bergii pelecanoides (King), crested tern; large, white, with back, rump, tail, etc. pearl-gray; crown black with crest. [For data on Micronesian birds, see Beker, Avifauna of Micronesia, U. of Kansas, 1951. Descriptions of sea birds: W, B, Alexander, Birds of the Ocean,/ REPTILES The Tanager Expedition, in 1923, collected specimens of two species of geckos and two species of skinks on Wake Island. These ean be distinguished as follows: A. Geckos have their body covered with small granules or minute scales; the top of the head is without symmetrical shields; the digits (toes) are dilated, and the pupil of the eye is vertical, GEKKONIDAB 1. The chin (below the mouth) has small scales, not forming shields; inner digits have a compressed distal phalanx which extends somewhat beyond its tip, like a claw; color pale with dark spots or markings. Lepidodactylus lugubris. 1', Chin-shields large for at least two rows; inner digits without a distal, compressed phalanx; a series of transverse plates under the tail. Peropus mutilatus. Panes B. Top of head with large symmetrical shields; body scales large, cycloid (thin, partially overlapping, with round outer edge, showing concentric growth lines); digits not dilated; pupil rounded. Skinks, SCINCIDAE .. 2, Eyelids well developed, movable; distinct light: ine down exact middle of upper surface from tip of snout to tail; bright blue-green tail. Emoia cyanura,. ~— 2', Eyelids indistinguishable, not movable; without a light-colored, mid-dorsal line. Cryptoblepharus or Ablepharus boutonii poecilopleurus. GECKOS Lepidedactylus lugubris (Dumeril & Bibron), the "sad" or mourning gecko, has a variable coloration, ranging from pale gray, brown, tan or pinkish on the back, with dark spots or blotches, and white or ~ pinkish lower surface, to very dark specimens, with or without a dark streak through the eye and along the side of the head. It is found from Indonesia to southeastern Polynesia, including Fiji, Samoa and - many of the low sandy islets of the central Pacific. It is the _ commonest gecko in Hawaii, and is widespread in Micronesia, It lays pear-shaped eggs with white, hard, thick shells, about 9 by 6.5 m., capable of falling and bouncing without .breaking. These are found cemented together or to a vertical surface, with clusters in cracks, under loose bark or stones, and against boards, Peropus mutilatus (Wiegmann), the stump-—toed gecko, gets its specific name from the fact that its skin is so thin and tender that one can scarcely catch a specimen without its being "mutilated" in some way. Struggles of the gecko against one's fingers are likely to tear rents in the skin, and the tail is easily broken off. The color of this species also is variable, changing from light to dark to agree with the habitat in which it is found. The under parts are more or ~ less tinted with yellow, which is more intense on the hind legs and belly. This species is widely distributed on Pacific islands from Mexico to the Philippines, especially on such low, reef islets as the Tuamotus and central Pacific atolls, also on higher islands such as the Society, Austral, Marquesas, Hawaii and Marianas. SKINKS Cryptoblepharus boutonii poecilopleurus (Wiegmann) formerly in Ablepharus, the snake-eyed skink, may be recognized most easily by the absence of eyelids and the absence of a light line down the exact middle of the back. Color is worthless in describing this species, for it may range from entirely dark to olive, brown, slaty, or even blue-green, uniform or having two to four lighter stripes on the back, but never an odd number (that is, with one in the exact center.) It is dis- tributed throughout Polynesia (including Hawaii), and-most.of Micro- nesia. It loses its tail easily, but may regenerate a new one. igs Emoia cyanura (Lesson), the azure-tailed skink, is recognizc4 easily by its sharply defined, light mid-dorsal line, which extends from the tip of the snout to the tail, which is bright blue. Most adults have the central light strip on an even, dark brown to black ground; some have three yellow or white stripes; in older specimens these may fade and almost disappear, Some authorities place this species in the genus Lygosoma, It occurs in many islands in the Facific, and is found throughout Micronesia. The egg measures 7 by li to 14 mm,; the skinks, 120 to 138 mm. in overall length. In 1952 this skink was fairly common about the island and Fosberg collected one on the ground in an open place. In March 1949, C. Morgan Holmes collected what is believed to have been a "cat snake" in a tree on Wake. This snake has been iden- tified as Boiga irregularis (Merrem), family Colubridae, subfamily Boiginae, It may be recognized by its large ventral plates, which stretch across the lower surface, several times longer than dorsal scales, It has 17 to 23 rows of scales, counted around the middle. The color is highly variable: yellowish, gray, olivaceous, bluish, greenish and brown above, uniform or spotted, It has been found in New Guinea, the Bismarck Archipelago, and the Solomon Islands, and probably reached Wake from one of these, FRESHWATER FISH Gambusia affinis Mosquito fish. In 1952 moSgquito fish were seen by F, Kk. Fosberg in ponds and bomb craters, as well as in a large cistern. Tiese had teen spread about 8 months earlier to the ponds and bomb craters by Mr. Fred Schultz, as a part of his campaign against mosquitoes, from a stock introduced into the concrete cistern on Peale Islet 4 years before from Hawaii, They were very numerous in October, 1952, although it was not clear what they lived on, since the mosquito population had been reduced almost to the vanishing point. INSECTS AND OTHER IAND ARTHROPODS The insects collected by the "Tanager Expedition" were reported in Bulletin 31 of Bishop Museum, by the writer and collaborators, 1926. Other notes on insects are giver by Aldrich, 1931, Bryan, 1948, Cresson, 1934, Hull, 1937, Jacot, 1928, 1929, Jordan 1939, Reeves, 1953, Rosen, Reeves and Aarons, 1948, Thompson, 1938, Usinger, 1937, 1941, 1946, 1949, and 1951, Van Zwaluvenburg, 1948, Wheeler, 1934, and Wiliiams, 1945 (see Bibliography for complete citations). More recent collections madé by Col. G. W. Bickness, A. T. Gramilini, F.C. Hadden, Dr. C. R. Joyce, Dr. M. L. Kenler, N. L, H. Krauss, J, P. Martin, and others, are being worked up by specialists and reported upon in the "Insects of Micronesia" series. In view of the fact thet only a portion of these reports have appeared tc cate, it would be premature to attempt to list the insects of Wake Island et. this time. es BIBLIOGRAPHY Air Weather Service. Weather conditions on a northern and a central Pacific air route. Air Weather Service Spec. St. 105-143: 1-9, 1950. Aldrich, J. M. New acalyptrate Diptera from the Pacific and Oriental regions. Proc, Hawaiian Ent. Soc. 7(3): 395-399, 1931. Tethina insularis n. sp. from Wake and Pearl and Hermes Reef, Pp. 336% Ayres, Pat A. Wake - the vegetable isle. Paradise of the Pacific 52(1): 13, January 1940. Discusses the hydroponic - "bath tub" gardens run by Torrey Lyons on Wake, Bailey, A. M. Notes on the birds of Midway and Wake Islands. Wilson Bulletin 63(1): 35-37, 1951. ~ Wake observations, in Stepping stones across the Pacific, by Alfred M, Bailey and Robert J. Niedrach. Museum Pictorial 3: 47-64, 1951. (Denver Museum of Natural History). a Bailey, J. W. and W. H. Harvey, Algae, in U. S. Exploring Expedition, Botany Cryptogamia 17: 153-192, Philadelphia, 1874. Includes Dictyosphaeria favulosa from Wake, p. 172. Bryan, E. H., Jr. and collaborators. Insects of Hawaii, Johnston Island and Wake Island. Bishop Museum Bull. 31: 1-94, 1926. Tabulates and describes insects; note on vegetation, p. &. Bryan, E. H., Jr. Insects of the Tanager Expedition [Abstract/. Pro. Hawaiian Acad. Sci,, lst annual meeting 1926. Bishop Museum Spec. Publ, 11: 31, 1926. Ephydrid fly new to Hawaii, Proc. Hawaiian Ent. Soc, 6(2): 279, 1926. Canace nudata Cresson from Wake, as well as Oahu, Lisianski I. High points on the clipper trip. Pan-Pacific Magazine 1(1): 37-43, Jan.—March 1937; map.. Popular observations on natural history of Wake I. American Polynesia and the Hawaiian Chain. 1-253, Honolulu, 1942; maps. ; Summary of Geography, natural history and history of Wake, pps elez—213. Flies on Wake. Proc, Hawaiian Ent. Soc. 13:221, 1948. Records Chrysomyia megacephala and Musca vicina (var. of M. domestica} as numerous on Wake, breeding on decaying animal and vegetable matter, laos Chilson, L. M. Fleas on Wake Island. Proc. Hawaiian Ent. Soc. 15(1): 1, March 1953. Fleas on Wake; Xenopsilla cheopis (Rothschild) collected May me 1951 in trash on warehouse floor. Christophersen, Erling. Vascular plants of Johnston and Wake Islands. Bishop Museum Occasional Papers 9(13): 1-20, 1931. Notes on vegetation and list of species, Tanager Expedition, 1923. Clark, Hubert Lyman. Echinoderms other than sea stars, in Marine zoology of tropical Central Pacific. Bishop Musewm Bull. 27: 89-111, 1925. Cresson, E, T., Jr. Descriptions of a new genus and species of the dipterous family Ephydridae XI. Trans. Amer. Ent. Soc. 60: 199-222, 1934. Canaceoides nudata from Lisianski and Wake Is. Cummins, G. B, Descriptions of tropical rusts. Torrey Bot. Club Uredo wakensis on Tournefortia. | Daniel, Hawthorn. Islands of the Pacific. New York, 1-228, 1943. Summary of the geography of Wake. Devereux, James Patrick Sinnott. The Story of Wake Island. 1-252, Philadelphia, 1947. Account of the mission on Wake, the attack on the island, md experiences in a prison camp, by the officer commanding the Marine garrison, Dickson, Donald /pseud. for Edwin North McClellan/. Wake Island. Paradise of the Pacific 51(4): 6, 25, April 1939. Historical note, Dixon, K. L. and Starrett, W. C. Offshore observations of tropical sea birds in the Western Pacific. The Auk 69: 266-272, 1952. Birds seen near Wake Island, 1945 and 1946. Drummond-Hay, H, A trip to Wake Island. China Journal 30: 333-339, 1939. Note on vegetation and ealeiadeied of vegetables. Edmondson, Charles H. Crustacea, in Marine zoology of tropical central Pacific. Bishop Museum Bull, 27: 3-62, 1925. Reports crustacea from Wake collected by Tanager Expedition, 1923. Findlay, Alexander George. Directory for the navigation of wens North Pacific Ocean. 3rd ed.: 1035-1036, 1886. Hydrologic and historic data. Fisher, W. K. Sea stars, in Marine zoology of tropical central Pacific. Bishop Museum Bull. 27: 63-38, 1925. Records Ferdina cancellata tylota n. sp., Linckia multifora, ne ee eee -18- Follansbee, W. Typhoon "Olive." 17 pp., 1952? Zauplicated/. Conditions on Wake during a typhoon. Fowler, Henry W. and S, C. Ball. Fishes of Hawaii, Johnston Island and Wake Island, Bishop Museum Bull. 26: 1-31, 1925. Records 95 species of fishes collected by Tanager Expedition. Freeman, Otis W. (ed.). Geography of the Pacific. New York (Wylie), 1-573, 1951. Notes on geography of Wake, pp. 6,27,36,65, 262,555,557. Gray, Asa, Botany, Phanerogamia. Part I. fpts. II and III never pub- lished/, U. S. Exploring Expedition 15: 1-777, Philadelphia, 18544 Record of plants observed and collected on eles ; Grant, Chapman. Frigate bird and the Laysan rail. Condor 49(3): 130, 1947. No fresh water on Wake, so Wake rail must have existed without water supply. Gregory, H. E, Report of the Director for 1923. Bishop Museum Full, TOs 2-36, 1922. Summary of Tanager Expedition; very little about Wake. - Grooch, W. S. Skyway to Asia. New York (Longman, Green), 1936. Account of the "North Haven" expedition which established bases on Midway, Wake and Guam for Pan-American Airways; pp. 83-124 and several photos on Wake. Hachisyka, M., N. Takatsukasa, S, Uchida, Y. Yamashina. Hand list of Japanese birds. Tokyo, 3rd and revised ed., 1-238, 1942. Lists birds recorded from Wake I, Hall, M. Journey to the end of an era. New York, 1-438, 1947. Memoirs of an air officer; notes birds of Midway and Wake, 372-375. Heinl, R. D., Jr. The defense of Wake.. U. S. Marine Corps Monograph, Washington, D. C., 1-75, 1947, Notes effect of military operations on vegetation of Wake. Official account of attacks with maps; brief survey of pre- war history; chronology, organization of defenders, text of Japanese surrender; notes on terrain and hydrology. Hull, F. M. A check list of the Syrphidae of . Oceania. "Bishop Museum Occasional Papers 13: 79-87, 1937. Records Ischiodon scutellaris from Johnston and ese Is. Jacot, Arthur P. New oribatoid mites. Psyche 35: 213-215, 1928, - Concerning the genus Neoliodes (Oribatoidea - Acarina), Am, Microscopical Soc, Trans, 48: 30-48, 1929. Udetaliodes hawaiiensis wakensis from Wake. anes Jacot, Arthur P, Some Hawaiian Oribatoidea (Acarina). Bishop Museum Bull, 121: 1-99, 1934; 16 pls. Ugetaliodes hawaiiensis wakensis, p. 10; Zetes bryani (Jacot), Wisi Were Jordan, K, On the constancy and variability of the differences between Old World species of Utetheisa. (Lepidoptera. Arctiidae). Novitates Zoolog. 41: 251-291, 1939. U. pulchelloides from Wake and other Pacific islands. Kaucher, D. Wings over Wake, San Francisco, 1-158, 1947. Illustrations show Wake vegetation. Kephart, Rodney. Wake, war and writing. New York, 1-84, 1950. King, Robert D. Index to the islands of the Territory of Hawaii... Honolulu, 1931, Brief summary of geography, position, history and jurisdiction of Wake, p. 27. Ileff, D. Ne Uncle Sam's Pacific islets. Stanford Univ., 1-71, 1940. Brief and not too accurate description; discovery and annexa- tion, 21-23, Mayr, E. Bird conservation in the southwest Pacific. Audubon Magazine 47: 279-282, 1945. Wake rail now extinct. Parr, Charles McKew. Over and above our Pacific, 1-274, 1941. Notes by a traveller stopping briefly at Wake on a flight across the Pacific, pp. 91-103; photo of pier, opposite p. 105. Pickering, C. Geographical distribution of animals and plants. JU. S. Exploring Expedition 19(2): 1-524, Philadelphia, 1876. Gives a list of plants on Wake's coral island, pp. 246-247, 489. Poole, A. J, and V. S. Schantz. Catalog of the type specimens of mammals in the U. S. National Museum, U. S. Nat. Museum Bull. 178: eee ee ee ee Compare: Tate, American Museum Nat. Hist. Bull, 68(3): 146, 1935, and U. S. Exploring Exped. 8: 47, 1848, orig. descr. Mus exulans Peale. Porter, H. P. The brave young birds of Wake Island. Elepaio 13: 47-48, 1953 e Young birds unharmed by typhoon Olive, Reeves, W. C. Possible recent introductions of mosquito vectors of human diseases in the central Pacific. Proc. 7th Pac. Sci. Congress 7: 371-373, 1953. Mosquitoes (Aedes aegypti and Culex quinquefasciatus) on Wake. -20— Reid, C. F. Overseas America: our territorial outposts. 1942. Brief paragraph and small map of Wake. Rosen, L., W. C. Reeves, and T, Aarons. Aedes aegypti on Wake. Proc. Hawaiian Ent. Soc. 13: 255-256, 1948. No mosquitoes on Wake prior to Dec. 24, 1941. Environmental conditions helping mosquitoes to live on island. Rothschild, Walter. Hypotaenidia wakensis n. sp. British Ornith. Club Bull. 13(99): 78, 1903. Original description of Wake Island rail. Rukeysen, Muriel. Wake island: a poem, New York, 1-16, 1942. Poetic impression of the island. Schmidt, 0, C. Verzeichnis der Meersalgen von Neu-Guinea und dem westlichen Oceanien, Hedwigia 68: 19-86, 1928. Includes record of Wake algae. Thompson, G. B. The Hippoboscidae (Diptera) recorded from the Eactete islands, Ent. Monthly Mag. 74: 14-17, 1938. - Olfersia spinifera from Laysan, Johnston and Wake Islands. Treadwell, A. L. Polychaetous annelids, in Marine zoology of tropical central Pacific, Bishop Museum Bull. 27: 113-119, 1925. Records 5 species from Wake, U. S. Navy, Chief of Naval Operations, Aerology Section, A climatic summary of Wake Island. Washington, D. C., 1-13, 1944. Also includes a brief geographical summary. U. S. Navy, Guide to the Western Pacific. CINCPAC-CINCPOA Bull. 126-44: 1-140, 1945. + Superficial geographical notes. U. S. Weather Bureau, (ope eeeeteet data - eavenae eis S747; 1949, and continuing.. e Summary of weather data. = Climatological data - national summary. Washington, D. OC. & Ashville, 1950 -, - Local climatological summary ~ Wake Island, Pacific, 1950- 1951 /continued as Local climatological data, from 1952/, Usinger, Robert L. Two new Pacific island species of Nysius (Lygaeidae-Hemiptera). Proc, Hawaiian Ent. Soc. 9(3): 439-442, 1937. Nysius picipes from Wake, on Sesuvium, Cordia, Portulaca, Boerhaavia, Sida. - The genus Oechalis (Pentatomidae, Hemiptera). Proc, Hawaiian Ent. Soc, ll: 59-93, 1941. Species from Wake, p. 74. od] = Usinger, Robert L. Heteroptera of Guam, in Insects of Guam. Bishop Museum Bull. 189: 11-103, 1946. Nysius picipes from Wake, - Wartime dispersal of Pacific islands Nysius (Hemiptera: lygaeidae), Proc, Hawaiian Ent. Soc. 13: 447, 1949. Nysius picipes Usinger on Wake I. - Heteroptera of the Marshall Islands. Proc. Hawaiian Ent. Soe, 14: 3053217 1951, Pentatomidae: Oechalia consocialis (Boisduval) and Lygaeidae: Nysius picipes Usinger on Wake. Van Zwaluwenburg, R. H. New species and new records of elaterid beetles from the Pacific, III. Proc. Hawaiian Ent. Soc. 13: 265-276, 1948. Conoderus pallipes (Eschscholtz) from Wake, p. 272. Vaughn, Charles. Observations on the birds of Wake Island. The Migrant 16(2): 26-28, 1945. Votaw, H. C, Wake Island. U. S. Naval Institute Proc. 67: 52-55, 1941. History of the atoll. Wetmore, Alexander, Bird life among lava rock and coral sand. Nat. Geogr. Mag. 48: 77-108, 1925. Account of the Tanager Expedition, 1923. Wheeler, W. M. Revised list of Hawaiian ants. Bishop Museum Occasional Papers 10(21): 1-21, 1934. Records: Monomorium floricola (Jerdon), M. pharaonis (Linnaeus), and Paratrechina (Nylanderia) bourbonica hawaiiensis Forel from Wake, Wilkes, Charles. Narrative of the U. S. Exploring Expedition during the years 1833-1842. 5: 1-591, Philadelphia, 1844. (several other editions). Includes visit to Wake, Dec. 20, 1841, p. 285; position, vege- tation, appeerance of having been submerged. Williams, F, X. Achaea janata (Linn.). Proc. Hawaiian Ent. Soc. 12233. ORS. This large noctuid moth found on Wake by Torrey Lyons; it fed on Cordia subcordata. Woodbury, D. O. Builders for battle. New York, 1-415, 1946. [Without specified authors/ Battle of Wake, Paradise of the Pacific 54(2): 10-12, Feb. 1942. Summary from official U. S. Navy release of attack on and cap- ture of Wake, -22— {Without specified authors/ A hydroponic farm on Wake Island. Science Supplement 87(2263 ) May 13, 1938. Pacific area communicable disease information service. Pacific Science Association Information Bulletin PS 52/6: 1-37, 1952. Includes table of diseases reported from Wake I, Wake Island's big blow, Typhoon levels Pacific outpost, life /magazine/ 28-29, Sept. 29, 1952: Illustrations. Wreck of the "Libelle". The Friend [Honolulu/ 17(9): 86, 1866. Account of wreck of a Bremen bark on Wake I,, March 4, 1866, ATOLL RESEARCH BULLETIN No, 67 Vegetation and flora of Wake Island by F, R. Fosberg Issued by THE PACIFIC SCIENCE BOARD National Academy of Sciences--National Research Council Washington, D. C. May 15, 1959 Vegetation and flora of Wake Island by Ba Ra Fosberg)/ Introduction Wake Island is a coral atoll lying north of the Marshall Islands at latitude 19917" North, Longitude 16°17! West. Although separated from Pokak, northernmost of the Marshalls by almost 550 kilometers, it is so similar to the Marshall Islands in most respects that it could almost as well be regarded as one of that group as an isolated island. It is a low flat atoll, its land area composed of three islets, Wake, Wilkes, and Peale, forming a V open to the northwest, surrounding a lagoon, the northwest end closed by a reef. A map of the atoll, as it was prior to the war, is given by Bryan (1942, p. 208), The south- east extremity is Peacock Point, named for one of the ships of the U. S. Exploring Expedition. Climate The climate is described by the U. S. Weather Bureau (1957) as follows: "The climate of Wake is maritime and is chiefly controlled by the easterly trade winds which dominate the island throughout the year. Occasionally during late fall, winter, and early spring, polar out- breaks reach the island and are marked by temperature drops of several degrees, increased cloudiness, and light to moderate showers of short duration, The winds during these outbreaks swing into the northerly directions and may reach gust velocities of forty miles per hour. After lI a few days the weather is modified and normality reigns again, It was during one of these outbreaks that the record low Cemperabure of . 64° F, was recorded. "Frequent tropical disturbances (low pressure systems, aloft and/ or surface) approach the island from the southeast quadrant during the late summer and early fall months. These systems bring about periods of light wind, high temperatures and humidities, and moderate to heavy rain showers. When the system is vigorous and close to the island, winds strengthen and showers may be prolonged to several hours dura- tion, Although typhoons are not usual, two have occurred at Wake since observations were begun in 1935: October 19, 1941, and September 16, 1952. Winds were up to 140 and 180 miles per hour respectively during the storms, "Clouds at Wake are predominately cumulus types with little dif- ference in amounts from day to night. High cloudiness, rare during the winter months, occurs in aaa with fall and summer tropical disturbances. ¥/Sotanist, U. S. Geological Survey, Washington, D. C. pm "Seemingly, showers, which account for most of Wake's precipi- tation, occur most frequently between midnight and sunrise; however, sufficient records have not been accumulated to be certain of the diurnal nature. The sky is seldom completely overcast, nor, on the other hand, is the sky often completely clear, "Thunderstorms are infrequent, but do occur along with the tropi- cal disturbances, however, hail has yet to be reported at the surface. Fog, also, is unknown to Wake." The mean annual precipitation, based on records from 1935 to 1941, 1947-1956, is given as 39,03 inches. Soils _ The soils are generally coral gravel or rubble of varied textures or in places largely coral sand, with here and there a slight coloring by humus. Under the Pisonia-Cordia forest a thin black friable humus layer lies on the coral. In one spot, the site of an old Jupanese garden, is a small area of dark brownish gray fine soil. Part of this is a cemented platform of brown rock which a spectrographic analysis by Helen Worthing and Catherine V:lentine, U. S, Geological Survey, showed to be phosphatic, with more than 10% phosphorus. No other consolidated rock was noticed except along the north coast of Wake Islet, where there is a long stretch of beach-rock below high tide level, overlain landward by beach conglomerate. Ve,etation The vegetation of Wake is low and the general appearance is gray. Most of the atoll is covered by an open scrub forest, mainly of Tourne- fortia, except where this has been cleared to make room for the air- field and other installations, Its aspect is similar to that of Pokak Atoll in the Marshalls, but the vegetation is somewhat more luxuriant and is definitely richer in native species, as well as in the many introduced ones. The first scientific information on the vegetation of Wake was published by Pickering (1876), an account of observations made there in 1841 by the naturalists of the U. S. Exploring Expedition, including Pickering, himself, This account notes "marks of a hurricane that had passed over the island from the Northeastward", absence of house- flies, absence of coconut palms, Pandanus, and Urticaceae, presence of rats and lizards, and that there were no certain indications of the visits of humans. No formal description of the vegetation is given but thirteen species of plants are listed, with brief notes on the occurrence of some of them. About the only difference from later accounts is that Pemphis is said to be the prevailing plant. Tournefortia is described as abundant. Tne olants are mentioned below under what seem to be their modern names. -A few had been listed by Gray (1854). The similarity of the lists of species recorded by Pickering and that given by Christophersen (1931) based on collections sas made 80 years later is striking. One species had apparently disappeared, and two additional native species were found by the later collectors. Four species had been introduced in the interval, presumably by the few casual visitors and the Japanese birdhunters, The vegetation of Wake was very well summarized by Christophersen (1931) who wrote up the botanical results of the Tanager Expedition of 1923, though he was not a member of that party. He quoted briefly from Wilkes, who visited the atoll in 1841, and more extensively from the field notes of E, H, Bryan who wa, a member of the Expedition. Judging from Bryan's observations, the vegetation was in fairly natural condi- tion at the time of his visit, though the presence of Portulaca oleracea, Nicotiana, Gossypium and, perhaps, of Ipomoea pes-caprae, suggests the influence of man, lLepidium o-waihiense, Sesuvium portulacastrum, and Heliotropium anomalum are three species found on Wake but lacking, so far as known, in the Marshall Islands flora. The last two of them are, however, known in the Marianas, and all three are found in Hawaii. Christophersen reports 16.species of vascular plants, 12 of them probably indigenous. In the 33 years since the collections listed by him were gathered, 26 species have established themselves, at least temporarily, in a wild condition and about 47 species of cultivated plants are now growing either in pots or with some form of protection. Additionally, at least five other native plants, Fimbristylis cymosa, Portulaca samoensis, a second species of Boerhavia, Abutilon albescens, and lepturus gasparricensis, have been found. The last may well have been Bryan's "broad-leaved bunch grass" which he apparently did not find in fertile condition. Bryan's plant could also have been Digitaria gaudichaudii which is similar when it lacks flowers and fruits. The vegetation is low, scrubby, and generally sparse, as would be expected on a rather dry atoll with a rainfall of about 1000 mm, a year, It is, however, subject to two sorts of climatic vicissitudes, both of which have occurred within the present decade, namely extreme droughts i and typhoons. In addition, between 1941 and 1945, Wake was almost to- tally devastated by military activity. The present paper reports observations and collections made during four visits, all short, made in October 1951, April 1952, July 1952, and October 1953. My appreciation is due to Mrs. Rubie Cox, Mr. Fred Schultz, Mr. T. Dudley Musson, Mr. B. H. Hall, and Mr. Edward H, Brown, for information, transportation, and other assistance during these visits, Judging from Bryan's description the original vegetation of the atoll was mainly a low woodland of Tournefortia (Messerschmidia), mixed here and there with Pemphis or Cordia, and with smaller areas of pure Pemphis forest and Cordia scrub, and with, on the north arm of Wa,e Islet, a dense forest of Pisonia, Cordia, and Tournefortia. Along the lagoon were areas of matted succulent Sesuvium and of a shrubby growth of Pemphis. Between the trees of the Tournefortia woodland was a sparse herbaceous and small shrubby growth of Lepturus, Heliotropium, Boerhavia, Lepidium, and Sida. Single small patches of Gossypium and a a Ipomoea pes-caprae occurred, the latter on the site of an abandoned Japanese camp dating from 1908. Ipomoea tuba was generally distributed, twining over rocks, buses, and trees. Because of the abundance of dead limbs and trunks noted by Bryan, it is suspected that a typhoon may have swept the island not long previously. During the latter part of 1941 the atoll was extensively fortified by a small American force, after which it suffered an extended attack ending in its capture by the Japanese on December 23, 1941. After this the fortifications were greatly extended by the Japanese, leaving scarcely any of the ground untouched, Ditches, tank traps, platforms, ruined buildings, gun emplacements, revetments, and underground installations were almost everywhere, with refuse of all sorts very abundant. According to T. D. Musson, in 1947 the vegetation generally over the island was about a foot high, with the exception of a few small areas. During the three years of Japanese occupation the island was subjected to almost daily bombardment by American planes and occasionally by ships. Anything edible in the vegetation was consumed by the starving Japanese, who also planted gardens in some places. By 1949, as shown by the excellent photos by Bailey and Niedrach (1951, pp. 47-62), the Tournefortia trees were already at least 3 m. tall. In 1952, the dominant vegetation over most of the atoll was a stand of Tournefortia 3-5 m, tall. Its recent origin was obvious from the lack of trunks more than 10 cm. thick and from the fact that it grew on embankments and piles of coral rubble that clearly dated from the war. Two areas, one near the east side of the lagoon, the other just east of the Trans-Ocean Airways compound, on the southwest limb of Wake Islet appeared to represent pre-war Tournefortia forest, with larger trees than usual. The latter of these areas had, in addition to the Tourne- fortia, fair sized Pemphis trees on the lagoon side that may have been part of the Pemphis forest seen by Bryan in 1923. A third area also seemed to have suffered rather less.than most others. This was a por- tion of the dense forest described by Bryan on the north limb of Wake Islet. A forest of Pisonia, with trees up to almost a meter in dia- meter, with some admixture of large tangled Cordia trees was still very prominent in 1952. The part of this area closest to the north beach, described by Bryan as having an impenetrable barrier of vegetation had, in 1952, only scattered herbs and bushes. Near Peacock Point were areas of a scrub growth of Cordia 1-2 m, tall, but probably grown up since the war. Scaevola sericea was mixed with the general Tournefortia woodland in some quantity only south of the air strip on the south leg of Wake Islet. Around most of the lagoon shore of Wake Islet was a broad belt of low Pemphis scrub, in places varied by Sesuvium flats. Though usually narrower, this belt in places reached 300 meters in width, In the openings among the Tournefortia trees, in addition to the scattered native herbs and small shrubs, there was a thin dry vegeta- tion of Euphorbia cyathophora and Cenchrus echinatus, obviously hard hit by the drought that had persisted from December to Avril. This had also affected noticeably most of the other plants. The Tournefortia had lost a good portion of its leaves, and, where growing on high embankments, in places looked quite dead. Pisonia and Cordia were more or less defoliated and Cordia in places dead at the tips of the branches. ihe Although in 1923 Bryan had noted only a single patch each of Gossypium and Ipomoea pes-caprae, in 1952 they were common in many places, especially around ruins of Japanese installations. Small grassy patches could also be seen here and there, some of them C.nodon dactylon which had been planted widely the previous fall but had not become very gene- rally established. Other patches were Paspalum vaginatum, which was apparently a recent introduction but well estabiished, A short visit in July 1952, after the drought had broken, showed a more luxuriant picture. The bare spaces between the trees were green with grasses and a luxuriant growth of Euphorbia cyathophora. The trees on high ground that looked dead in April were still dead in July, but the others, mostly in lower places, looked far more luxuriant than on the previous visit. During the first three visits relatively little planting of orna- mentals or garden ovlants was observed. A few plants, including coco- nut, Crinum, Dieffenbachia, Polyscias guilfoylei, and Casuarina were growing behind shelters and windbreaks around the dispensary, and a few others in pots and boxes of imported soil. At the TAL (Transocean Air Lines) Compound were more extensive attempts at gardening. Around the buildings were one coconut tree and one Pandanus tree, both at least 2m. tall, and small plants of Cesuarina, Ficus carica, Ficus rubiginosa, Euphorbia pulcherrima, Thespesia, Terminalia, Cucurbita pepo, and Catharanthus, as well as broomcorn. A number of other ornamentals apd maize were growing in pots. The method of vegetable gardening used most successfully was in top soil niled about 15 cm. deep, surrounded by boards, watered well, and shaded with leafy branches of Tournefortia. In such frames were growing successfully tomatoes, lettuce, broccoli, parsley, dill, onions, beans, radishes, and cucumbers. In September of 1952 a violent typhoon lashed the atoll. Winds reported as up to 140, or even 160, miles per hour, with gusts to 160 or 180, were experienced. A great deal of wind-driven flying gravel was observed by eyewitnesses, though it stopped as soon as water covered an area, Little topographic change resulted, as few waves actually swept the island. Some areas were covered, though, by water backed up from the lagoon. It was said by those who experienced the typhoon that all leaves, as well as some bark, were stripped from the trees apparently by flying gravel. The leaves, however, began to reappear after about 10 days. Very few trees were uprooted, though many were badly broken. The most conspicuous damage was that done by flying gravel. A two day visit was made by me in October, 1953, especially to observe what recovery from the effects of the typlioon had taken place. The following account is quoted, almost verbatim, from notes made at that time: "The present condition of the vegetation, after an unusually wet season, is one of notable luxuriance compared with Asril 1952 but the trees present a much more ragged appearance, with many dead branches and some tendency to lean and to be partially broken in one direction or another, depending on the locality on the island. In only a very few areas have many trees been actually uprooted, but enormous quan- tities of branches have been broken off and tend to make progress =6= through the undergrowth slow and difficult. Tangled masses of wreckage, twisted iron, bent corrugated sheets, and other debris are also strewn through the brush, "The dead branches of trees are usually ohes from which significant areas of bark have been stripped or ones that have been broken. Tourne- fortia and Pisonia trees suffered most, some of the latter being blown down. Cordia apparently resisted the wind more effectively, as relatively little dead wood is in evidence on it. The Cordia-Pisonia forest seems still to be quite intact, though some trees were blown down, especially Plconia, which usually takes root where it touches the ground. The whole is quite tangled, choked with branches and fallen trees. "The Tournefortia forest is also in most places still in reasona- bly good condition, but in a few places, such as east of the TAL Com- pound on Wake Islet and areas on the middle and east thirds of Peale... Islet, it presents a very battered appearance, indeed, most of the trees being broken. The Tournefortia trees on high embankments near the TAL Compound that were dead in 1952 are now practically gone. "There are now almost no areas of large Pemphis left, but it is difficult to tell if they were destroyed by the typhoon or by clearing to make roads and to put in other installations. The low brushy Pemphis at the east end of the lagoon was apparently killed back somewhat almost everywhere, hut has largely recovered. Dead branches and top portions of plants are still very much in evidence. "Two of the three patches of original forest that were still quite distinguishable in 1952 could not even be located this time. They apparently suffered more than the lower second growth. That near the east end of the lagoon, however, may have been destroyed by clearing. The Cordia scrub near Peacock Point is much greener and more luxuriant than before. "The general effect of the typhoon in many places seems to have been to destroy or greatly inhibit the growth of young woody plants which were invading bare areas, so that many pioneer habitats have been maintained. Tnis has encouraged a great spread and outstanding luxuriance of first- stage pioneers like Cenchrus echinatus, Cenchrus brownii, Boerhavia diffusa, Setaria verticillata, Euphorbia cyathophora, Eleusine indica, and Fimbristylis cymosa, and to a somewhat lesser extent, Portulaca oleracea (which seems to be hybridizing with P. lutea), Digitaria insularis, Euphorbia glomerifera, and others. "Small but vigorous patches of Cynodon dactylon, Paspalum vaginatun, Cyperus rotundus, and Euphorbia prostrata are found here and there. Gossypium religiosum is local but widespread, and very uniform in appearance. Ipomoea pes-caprae, Portulaca lutea, Heliotropium anomalun, and Sida fallax are about as common as before. Heliotropium is colonizing actively the b:re gravel flats and beaches on the west ends of Peale and Wilkes islets. Lepturus repens seems more widespread and common than before, but probably this may be due to the much wetter season. Lepturus gasparricensis, not seen before, forms a colony of a ere ne oe perhaps 40-59 clumps near the west end of Peale Islet. fhe "A few scattered Scaevola bushes were seen in the brush south of the runway, and one at Peacock Point, but this species seems rarer than in 1952 and is completely gone from around the Terminal. "Recently cleared areas come up to a thick Tournefortia scrub rather promptly. The antenna-area of the LORAN Station must be cleared every month or so. The Tournefortia sprouts grow up to about 0.8 m, in that time. The ground is dragged with a steel I-beam to break off the shoots. If this is not done the growth slows down so that they are about 1 to 2.5m. tall after 3 or 4 months. "The cultivated plants around the TAL Compound are growing well-- Ficus carica is up to 1.5 m. tall and bearing fruit, Terminalia catappa is 1.5 m. tall and healthy. Carica papaya, Catharanthus roseus, Casuarina equisetifolia, and Hibiscus hybrids are in good condition. Two forms of Cepsicum are also in “healthy condition and fruiting abun- dantly. Buuhinia sp. sp. is chlorotic and badly eaten by insects. Canta- loupes are growing well and flovering and fruiting around the Communi- cations Center, but are slightly chlorotic on exposed sides of the build- ing. These formerly had to be pollinated by hand, but are now being visited by bees from a swerm recently brought from Honolulu. A resident reported that a single tomato plant had grown and fruited near the garbage dump. Casuarina has been planted and is growing at several places, and coconuts are being planted around all building areas. Crinum is still surviving and growing at the Dispensary. "CAA personnel are much interested in planting around the buildings and are establishing an "arboretum and nursery" where they intend to start and grow vlants. Tney have many cuttings and seedlings, obtained from Honolulu and Midway. Many potted nlants such as Hymenocallis littoralis, Coleus scutellarioides and Casuarina are waiting to be planted out. Many of these will probably + do well enough during this wet spell, but some may fail in a drought. Tney.are being watered with brackish water from wells which is now rather fresh but which will be increasingly saline when the weather is drier. This is especially so since the wells are closely spaced, 6 or 8 of them in a row 15 m. or so apart, thus drawing excessively from one small part of the fresh-water lens." These observations are of a rather random nature and it is a pity that they are not more up to date, The records of the presence and condition of certain exotic plants at a definite time may be of interest in later interpretation of the vegetation. From observations to date the development of vegetation in dry coral islands such as this. Un- disturbed it may gradually give place to Pisonia or Pisonia-Cordia forest. With extreme fluctuations in rainfall, and especially with ocersional typhoons this process is greatly retarded. Since the Tourne- fortia colonizes successfully bare sand and gravel, Tournefortia wood- land may immediately succeed itself if destroyed by storms or other causes, Where the salinity is too high, Pemphis and Scsuvium form the normal vegetation. = The Vascular Flora Plants have been collected on Wake Island by the U.S. Exploring Exdedition in 1841, Pollock and his associates on the Tanager Exzedition in 1923, R. J. Branckamp in 1936, Mr. Torrey Lyons in 1940, F. R. Fosberg in 1951, 1952 and 1953, and by Dr. and Mrs. John 2. Gaston, in 1952 and 1953. Possibly other collections have been made of which we have no record. Vascular plants have been recorded from Wake Island by Gray (1854), Pickering (1876), Bryan (1926), Christophersen (1931), Drummond- Hay (1939), Bryan: (1942), and Fosberg (1955 a, b). In the present paper I have tried to collect together the records from all of these accounts, and to cite specimens from my ow collections and those of others if I have examined them. A few species have been added from what seem to be reliable sight records even though no specimens were collected. Older records, made under names other than those used in the present list, are referred to their presumed modern equivalents, but with the older name given in parentheses, Tne fact that a species was recorded in my 1955 list is not mentioned here unless there has been a name change. Unfortunately I have not been able to examine any of the Gaston collec- tions, and only a few of the Pollock, Branckamp, and Lyons ones, all of which are housed in the Bishop Museum, The collections cited below are deposited in the U, S. National Herbarium except those of Pollock, Branckamp, and Lyons. Svecies believed to have been introduced by man, rather than naturally, are indicated by an asterisk. PANDANACEAR *PANDANUS TECTORIUS Park, Pandan, or Screw Pine. A single tree 2m. tall seen in 1952, planted in housing area, Characterized by stilt roots, thick, irregular branches, and linear leaves several feet long, with spiny margins and midrib. GRAMINEAE *CENCHRUS BROWNII R. & S, Sand Bur, or Bur Grass. A pioneer grass with bur-like fruits, similar to and often con- fused with the next, but the fruits are more numerous in narrower, longer, more compact spikes. Fosberg 34929. *CE‘CHRUS ECHINATUS L, - Sand Bur, or Bur Grass. A very common early pioneer weed, quickly covering open places; the fruit is a most annoying bur which scratches the skin and clings to clothing. Hosters 33624. ¥CHLORIS INFLATA Link Finger Grass. A common pioneer grass with several bristly purple spikes as- cending from the top of the stem. Fosberg 34920. *#CYNODON DACTYLON (L.) Pers. Bermuda Grass. Certainly introduced and established since Bryan's visit in 1923, though possibly before the war; much more of it planted in 1951, accord- ing to local information. Forms dense mats of rhizomes, flowering stems slender, erect, with 3-4 radiating slender spikes. Fosberg 34931. (Bryan 1942, p. 213.) | *XDACTYLOCTENTUM AEGYPTIUM (L.) Reich. ~° Crowfoot Grass, : A tufted depressed grass with 4-5 stiff, pointed, thick flover- spikes radiating from the tips of the stems. Comnon in open situations. Fosberg 33617. “DIGITARIA CILIARIS (Retz.) Koel. Tropical Crab Grass. A slender spreading grass with very slender divergent spikes of flowers, Not abundant but found on Peale Islet repeatedly since 1940. Branckamp s.n. in 1936; Lyons 10; Fosberg 34936. DIGITARIA GAUDICHAUDII (Kunth) Henr. A rare strand grass, otherwise known only from the Marianas and possibly introduced from there, with broad leaves and close clusters of stiff spikes. This may be the "broad-leaved bunch-grass" mentioned by Bryan (Christophersen 1931, pp. 7-8), though it also may have been lepturus gasparricensis, as vegetatively they look much alike. Lyons oe tbryan) VOL?) 5, 213. ) *DIGITARIA INSULARIS (L.) Henr. Sour Grass. An aggressive pioneer grass, forming erect bunches, the inflores- | cence a cluster of somewhat nodding, silky spike-like racemes. This : has been known for a long time in Hawaii and has, since the war, appeared in a number of Micronesian islands, probably transported with war materials, Fosberg 34476. (Fosberg 1955a, as Tricachne insularis (L,) Nees) *ELEUSINE INDICA (L.) Gaertn. Goose Grass. A very common grass around dwellings and installations, forming rather coarse spreading tufts, inflorescence a cluster of flattened spikes spreading from the end of the stem, Fosberg 33623. *ERAGROSTIS AMABILIS (L.) W. & A. Love Grass. A spreading delicate grass with very fine open flower clusters, usually dark purplish; common in open places, especially gardens and around dwellings and installations. Fosberg 34459. *ERAGROSTIS POAEOIDES Beauv. ex R. & S, A rare tufted prostrate grass, similar to the preceding but with pale, longer spikelets of flowers; found once at the end of the airstrip, not found previously in the Pacific Islands. Fosberg 34458. LEPTURUS GASPARRICENSIS Fosb. A coarse bunchgrass with broad leaves ead rather coarse jointed flower spikes, occurring in a restricted area toward the west end of Peale Islet, known only from here and from Pokak Atoll, in the Marshalls. Possibly the "broad~leaved bunchgrass" mentioned by Bryan (Christophersen 1931). Fosberg 34940. (Fosberg 1955a, as lepturus sp.; Fosberg 1955b, pe 299) LEPTURUS REPENS var. SEPTEYTRIONALIS Fosb. Bunch Grass. The common bunch grass, forming erect tufts with fine leaves and long slender spikes that become disjointed when ripe. JU. S. Expl. saped. s.n. in 1841; Branckamp s.n. in 1936; Lyons 10; Fosberg 33627, 34462 34941, 34942, 34943, 34944, 34950. (Pickering 1876, p. 247, as le turus ; Christophersen 1931; Bryan 1942, ». 213; Fosberg 1955a, all three as Lepturus repens (Forst. ) R. Br.; Fosberg 1955b. ) ee -10=- *PASPALUM VAGINATUM Sw. Salt Grass. A wiry creeping, mat-forming grass, somewhat coarser than bermuda grass, forming patches locally. Fosberg 34932, 34949. *SETARIA VERTICILLATA (L.) Beauv. Bristly Foxtail. A loose somewhat spreading grass with bristly pointed flower ela. ters that, when ripe, cling to clothing and thus cause the seeds to he scattered about; occurs very locally in semi-shaded areas. Fosberg 34470. *SORGHUM DOCHNA var. TECHNICUM (Koern.) Snowd. Broomcorn. Persisting, in 1952, very sparingly around old Japanese garden-sites, also seen at TAL Compound. *ZEA MAYS L, Maize, or Indian Corn. Grown in drum of soil at TAL Compound in 1952; reached maturity but Was very spindly and with only abortive ears. UNIDENTIFIED GRASS Pickering, 1876, p. 247, reported "Gramen incert., (No. 1; compare Tarawan coral-islands). Single stemmed, and large-leaved; resembling young Phragmites." There is nothing at present on Wake that would even remotely answer to this description. CYPERACEAR FIMBRISTYLIS CYMOSA R. Br. A low tufted bright green grass-like plant, with open brownish clusters of flower-spikes on slender stems rising above the leaves; occurring very commonly from extreme pioneer situations to wooded placés, It was surprising that this was not mentioned from Wake in the earlier descriptions, as it is most likely native and usually occurs abundantly wherever it is found. Fosberg 33619, 34457, 34478. (The Wake Island plant belongs to the form recently segregated from F. cymosa as F. atollensis St, Jonn. It is, however, too close to what has been called F, spathacea Roth, and Kern, eminent student of sedges of the Malaysian area regards the whole complex as belonging to one species. Probably the plants of the Pacific atolls and seashores may constitute a good variety, distinguished from F, spathacea by smooth nuts and from F. cymosa by two stigmas and plane-convex nuts, but as yet this study is unpublished and it seems best to call the whole group simply F. cymosa. ) PAIMAE *COCOS NUCIFERA L. Coconut Palm. A few not too healthy seedlings which were planted in the shelter of the buildings were eliminated by the typhoon, but many have been re- planted. ARACEAB *CALADIUM sp. Mentioned by E, H. Bryan (m.s.) as planted about houses. ih *DIEFFENBACHIA sp. Dumb-C-ne. One potted plant seen in 1952, *SCINDAPSUS AUREUS (Lind. & Andre) Engl. Pothos. One or two potted plants seen in 1952. PONTEDER TACEAE *EICHHORNIA CRASSIPES (Mart. & Zucc.) Salms-L.ub. Water Hyacinth. One plant seen in a large cistern in 1952, said to be the only one remaining of a large number, the rest of which had been taken by the inhabitants to decorate aquaria, BROMELIACEAE *ANANAS COMOSUS (L.) Merr. Pineapple. A rooted crown seen growing in a box of soil in 1952. AGAV ACEAE *CORDYLINE TERMINALIS (L.) Kunth Tay Seen growing in pots in 1952, *SANSEVIERIA ROXBURGHIANA Schultes Bowstring Hemp. Seen growing in pots in 1952, AMARYLLIDACEAE *ALLIUM sp. Onion. Seen growing in protected garden at TAL Compound in 1952, *CRINUM sp. Crinum Lily. Seen growing at dispensary in 1952 and 1953, but not flowering. May be distinguished from the spider lily by the spiral arrangement of the rosette of lance-shaped leaves and the lack of a web between the stamens. *HYMENOCALLIS LITTORALIS (Jacq.) Salisb. Spider Lily. Growing in pots in nursery in 1953, waiting to be planted out. _ The leaves of this are packed in a tight row, rather than spirally as in Crinum, and are linear and blunt at the apex, CASUARINACEAE *CASUARINA EQUISETIFOLIA L. Ironwood, Planted, doing well around TAL Compound and buildings elsewhere in 1953; a tree superficially resembling a pine, but with "needles" jointed and bearing rings of tiny scales. eee MORACEAE *FICUS CARICA L. Fig. Planted around the TAL Compound, 1.5 m. tall and fruiting in 1953; recognizable by its roughish leaves with finger-like lobes. Fosberg 34466. *FICUS RIBIGINOSA Desf. ? One small plant growing at the TAL Compound in 1952. Fosberg 34467. POLYGONACEAE *COCCOLOBA UVIFERA (L.) L. Sea-Grape, A coarse shrub with large round leathery leaves and edible grape- like fruits; rare but apparently naturalized in woody vegetation on Wake and Peale Islets, where it apparently survived the typhoon, as the clump on Peale Islet was:2.5 m. tall in 1953. Fosberg 34471, 34935. NYCTAGINACEAE BOERHAVIA DIFFUSA L. A prostrate mat-forming plant with long reddish stems radiating from a thick fleshy root-crown, leaves obtuse, pale beneath, flowers pink; common generally. Fosberg 33626, 34453. (Pickering 1876?; Drummond-Hay 1939; Bryan 1942, p. 213.) BOERHAVIA sp. Similar to the above but with pointed leaves and white flowers, common in open places on sand. A distinctive species, known from Wake and the Northern Marshalls, perhaps also from Johnston I., usually referred to B. diffusa, but probably an undescribed species. Fosberg 34454, 34947. (Pickering 1876?; Christophersen 1931, as B. diffusa L.; Drummond-Hay 1939; Fosberg 1955a, as Boerhavia, form with leaves as in B. mutabilis R. Br.) *BOUGAINVILIEA SPECTABILIS Willd. Bougainvillea. Mentioned by E, H, Bryan (m.s.) as a garden plant. PISONIA GRANDIS R. Br. Puka. One of the principal species in the patch of low forest on the north- west leg of Wake Islet, a tree with thick white trunk, very soft wood; large green leaves, clusters of greenish flowers and club-shaped spiny glutinous fruits. Fosberg 34472. (Bryan 1926, p. 8; Christophersen 1931 as Ceodes sp.; Drummond-Hay 1939; Bryan 1942, p. 213.) AMARANTHACEAE *AMARANTHUS DUBIUS Mart. Amaranth Pig¢weed. Erect herb with slender nodding spikes of tiny green flowers; occasional along roadsides in scrub on Peale Islet. Fosberg 34938. Uiee *AMARANTHUS GRAECIZANS L. Amaranth Pigweed, A depressed, small-leafed annual weed, collected on Peale Islet. Branckamp s.n. in 1936.: (Bryan 1942, p. 213.) *AMARANTHUS TRICOLOR L. Amaranth Pigweed. Erect plant, similar to A. dubius but spike not nodding; persisting around old Japanese garden site. Fosberg 34452. *AMARANTHUS VIRIDIS L. Chinese spinach. Sprawling herb with small clusters of tiny green flowers; persisting around old Japanese garden site, Fosberg 34446. ATZOACEAE SESUVIUM PORTULACAS™RUM L. Seaside Purslane. A prostrate very fleshy plant with glossy green leaves, reddish stems, and rose-.ink flowers; common generally in low saline places, forming a continuous mat just above ii level inside the passage at east end of Peale Islet. Fosberg 34461. (Gray 1854, p. 142; Pickering 1876, p. 247; Christophersen 1931, p. ails Pe umnon ey 1939, Bryan 1942, pe 213, as pickle-weed, ) PORTULACACEAE PORTULACA LUTEA Sol. Giant Purslane. Very fleshy herb with thick stems grayish near the base, ascending to erect, leaves broad at apex, flowers large, yellow, with many stamens; common generally in open places in scrub. Fosberg 34468. (Gray 1854, p. 139, as P. oleracea; Pickering 1876, p. 247, as Portulaca; Christo- phersen 1931, p. 11; Drummond-Hay 1939; Bryen 1942, p. 213.) *PORTULACA OLERACEA L. Purslane. Similar to P. lutea but depressed, more slender, brownish green, stems not grayish at base, flowers small, with 12-17 stamens; occasional on bare ground around terminal and elsewhere, more abundant after the typhoon, apparently hybridizing with P. lutea. The Pacific plants have different seeds from the European plants and may be a distinct rative species, This has been in Wake at least since 1923, when Pollock found it. Fosberg 33620. (Christophersen 1931, p. 12; Drummond-Hay 1939; Bryan 1942, p. 213.) PORTULACA SAMOEMSIS v. Poelln. A prostrate plant with tuberous root, narrow pointed fleshy leaves with copious tufts of long hair at the base, and bright yellow flowers; rare in open places, This is a native Pacific Island species but possibly might be introduced on Wake, as it was not found earlier. Fosberg 34460. CRUCIFERAE *BRASSICA OIERACEA var. ITALICA Plenck. Broccoli. Grown in sheltered raised beds of soil, watered well, at TAL Com- pound in 1952. ye LEPIDIUM O-WAIHIENSE C, & S. Scurvy Grass. A small rather stiff erect cress-like plant, noted by Bryan in 1923 as forming part of the undergrowth on the western parts of Wilkes and Peale Islets, and collected on all islets by Pollock, but in 1953 seen only in a flourishing colony on steep banks around "Lake Peale", a pond on Peale Islet. Fosberg 34930. (Pickering 1876, p. 247; Christophersen 1931, p. 13; Drummond-Hay 1939; Bryan 1942, p. 213.) *RAPHANUS SATIVUS L. Radish. Grown in raised sheltered beds at TAL Compound in 1952. CRASSULACEAE *SEMPERVIVUM TECTORUM L, Hen-and-Chickens. Seen in pot in housing area in 1952. *KALANCHOE PINNATA (Lam.) Pers. Air Plant. This, with two other unidentified species of Kalanchoe seen growing in pots in housing area in 1952, LEGUMINOSAE *BAUHINIA sp. Orchid Tree, Seen in pot at TAL Compound in 1952, still there in 1953 but chlorotic and badly eaten by insects; may be recognized by its leaves which are divided into two lobes at the apex. *PHASEOLUS VULGARIS L. Bean. Growm in raised, sheltered garden bed at TAL Compound in 1952; - mentioned by Drummond-Hay, 1939, as growing in hydroponic cultures. EUPHORBIACEAE *CODIAEUM VARIEGATUM (L.) Bl. ~ Croton. Seen in pot at TAL Compound in 1952. *EUPHORBIA CYATIOPHORA Murr, Dwarf Poinsettia. A herb up to 0.5m, tall, with milky sap, top leaves, under the flower clusters, partly brilliant scarlet; very abundant in open places, mostly almost dead in dry period in April, 1952, but flourishing after rains in July. Fosberg 34474. (Commonly but incorrectly referred to as Euphorbia heterophylla L., which is a related species much less common in the Pacific.) *EUPHORRIA GLOMFRIFERA (Millso,.) Wheeler Spurge. A delicate erect milky herb or slender shrub, arching at tips, with tiny white flowers; comson around Terminal in 1951, had spread widely in open ground and wag common after the typhoon in 1953. Fosberg 33618, 33622. (Commonly but apparently incorrectly called E. hypericifolia L.) igs *EUPHORBIA HIRTA L. Hairy Spurge. A small arching milky hairy herb with pointed leaves and clusters of tiny white flowers; local on bare ground around Terminal. Fosberg 33628, *EUPHORBIA PROSTRATA Ait, Prostrate Spurge. A tiny prostrate purplish plant with milky sap; rare in open gravel on Wake Islet in 1952, occasional in bare spots on Peale Islet in 1953, Fosberg 34455, 34939. *EUPHORBIA PULCHERRIMA Willd. Poinsettia, Small plants seen at TAL Compound in 1952. *PHYLLANTHUS AMARUS Schum. & Thonn. Reported as a weed by E. H. Bryan (1942) (as P. niruri L.), pro- bably P, amarus but specimen not seen. MALVACEAE ABUTILOM ALBESCENS Miq. Shrub 1 m. tall with velvety leaves, orange flowers and hairy pa-.. pery fruit segments; one plant seen in 1952 near old Japanese installation. Fosberg 34451. (Generally referred to A, indicum (L.) Sweet, from which it differs in aspect, much larger flowers, and different fruit, probably actually closer to A, asiaticum (L.) D. Don.) *GOSSYPIUM RELIGIOSUM L. Wild Cotton. A spreading reclining shrub with grayish leaves, light yellow flowers, and diminutive cotton-producing fruits. Bryan reported one clump in 1923, but in 1952 fairly common especially around old Japanese installations on Wake Islet. Native in the Pacific, but rare and known from very few specimens, probably but not certainly an introduced plant on Wake, first found there in 1923. Fosberg 34469. (Christophersen 1931, p. 13, as G. hirsutum var. religiosa W:tt; Drummond-Hay 1939; Bryan 1942, p. 213, also as G, hirsutum var, religiosa. ) *HIBISCUS (ornamental hybrids) Hibiscus. seen in 1953 at TAL Compound, growing very well. SIDA FALLAX Walp. Tlima. A gray-leafed shrub with dull orange flowers; scattered generally in sparse scrub and in openings, about as abundant after the typhoon as before. Fosberg 34945. (Gray 1854, p. 161; Pickering 1876, p. 247; Christophersen 1931, p. 13; Drummond-Hay 1939; Bryan 1942, p. 213.) *THESPESTA POPULNEA (L.) Sol. ex Correa Milo, or Portia Tree. Seen in 1952 planted at TAL Compound. PASSIFLORACEAE *PASSIFLORA sp. Water Lemon, "Some kind of water lemon seen in the 1940's" according to E. H. Bryan (m.s.). =t6= CARICACEAE SCARICA PAPAYA -L, Papaya. Seen growing at TAL Compound, in 1952 said to reach only a Jew dm. in height and-then to die, but in 1953 appeared in good condition. CUCURBITACEAE *CUCUMIS MELO L. Cantaloupe. Growing well and fruiting around Communications Center in 1953. Formerly had to be hand pollinated, but later a swarm of bees wes brought from Honolulu and these visited the blossoms regularly, according to local information. *CUCURBITA PEPO L. Pumpkin. Growing in raised sheltered g-rden plot at TAL Compound in 1952; a squash is also mentioned by Drummond-Hay (1939) as growing in hydro- ponic culture. LYTHRACEAE PEMPHIS ACIDULA Forst. f. A closely branched shrub or small tree with very hard wood, small leaves with an astringent taste, and white flower, that is one of the dominant plants on certain areas of the island, forming scrubby forests locally east and south of the lagoon and scrub around the lagoon margin. . (Gray 1854, p. 601; Pickering 1876, p. 247; Bryan 1926, p. 8; Christophersen 1931, p. 14; Drummond-Hay 1939; Bryan 1942, p. 213.) COMBRET ACEAE- *TERMINALIA CATAPPA L. ~ _ Indian Almond. A tree with large obovate leaves which turn red before falling and somewhat flattened egg-shaped fruits with a keel along one side and 4 point; seen planted around the TAL Compound in 1952, It survived the typhoon and was growing well in 1953. MYRTACEAE *EUCALYPTUS CITRIODORA Hook. Lemon-Scented Eucalyptus. Seen growing in a pot at TAL Compound in 1952. ARALIACEAE - *POLYSCIAS GUILFOYLEI (Cogn. & March.) Bailey Hedge Panax. Seen growing in pot at Dispensary in 1952. Sid to grow well, reaching a height of two feet in calm season, only to be cut down almost to ground by dry winds during windy weather. To fi UMBELLIFERAE *ANETHUM GRAVEOLENS L. Daady Seen growing in raised, sheltered garden plot at TAL Compound in 1952. *APIUM PETROSELINUM L, Parsley. .. Seen growing in raised, sheltered garden plot at TAL Compound in 1952. SAPOTACEAE *CHRYSOPHYLLUM CAIMITO L.? Star Apple. Seen growing in pot at TAL Compound in 1952. APOCYNACEAE *xCATHARANTHUS ROSEUS (L.) Gi. Don Periwinkle. A small erect herb 3-4 dm. tall, with opposite oblong leaves and terminal clusters of flowers, varying on different plants from white to magenta; first reported in 1940 by Torrey Lyons, seen growing in shel- tered places around TAL Compound and Dispensary in 1952 and 1953, rare in open places elsewhere in 1952 but apparently not so widespread in 1953. Fosberg 34456. CONVOLVULACEAE #IPOMOBA BATATAS (L.) Poir. Sweet Potato. - Said to have grown luxuriantly at the Dispensary when sheltered from wind, but parts exposed to wind were killed instantly. Later insects ate the whole patch. Also seen in 1952 persisting, but rare, around old Japanese garden site on Wake Islet. Fosberg 34449. *TPOMOEA PES-CAPRAE ssp. BRASILIENSIS (L.) v. Ooststr. Beach Morning-glory. A prostrate extensively spreading creeper with leaves with two rounded lobes at the tip and rose~purple flowers. Bryan reported one patch, near an old Japanese encampment near the east end of Peale Islet in 1923 (Christophersen 1931, p. 8); it is now common locally in many parts of Wake and Peale Islets. Although commonly regarded as native in most tropical strand habitats, it is probably introduced on Wake Atoll. Fosberg 34473. (Christophersen 1931, p. 14; Bryan 1942, p. 213, both as Ipomoea pes-caprae (L.) Roth.) TPOMOBA TUBA (Schlecht. ) Don Moon Flower. An extensive creeper and climber with heart-shaped leaves and large white flowers which open at night; very common generally. Fosberg 33614. (Pickering 1876, p. 247, as "(Ipomoea) turpethum?"; Christophersen 1931, p. 14, as Calonyction grandiflorum (Jacq.) Choisy; Drummond-Hay 1939, as morning-glory; Bryan 1942, p. 213, as Ipomoea grandiflora. } we BORAGINACEAE CORDIA SUBCORDATA Lam. Kou, A shrub or small tree with ovate or elliptic rather large leaves and showy brilliant orange flowers; dominant or codominant in two vege- tation types, the Pisonia-Cordia forest and the Cordia scrub, both on Wake Islet, occasional elsewhere. Fosberg 34480. (Pickering 1876, p. 247, as Cordia; Christophersen 1931, p. 14; Drummond Hay 1939, Bryan 1942.) HELIOTROPIUM ANOMALUM H. & A. Beach Heliotrope. A low silvery leafed fleshy herb or dwarf shrub with dense clus- ters of yellowish-white very fragrant flowers; common generally in bare places and openings in scrub, one of the first pioneer plants to appear on beaches. Fosberg 33625. (Pickering 1876, p. 247, as "(Pentacarya? No, 2, or compare) Tournefortia"; Bryan 1926, p. 8; Christophersen 1931, p. 14; Drummond-Hay 1939; Bryan 1942, p. 213.) TCURNEFORTIA ARGENTEA L. f. Tree Heliotrope. A low rounded tree or shrub with fleshy obovate silvery or frosty appearing leaves and clusters of small fragrant white flowers; dominant in the vegetation of the largest parts of all three islets. Commonly called "scaviola" on the island, which seems to be a corruption of Scaevola, the name of another plant of vaguely similar appearance. This usaye should be discouraged as it is likely to be confusing. Fosberg 33625. (Pickering 1876, p. 247, as Tournefortia; Bryan 1926, p. 8; Christophersen 1931, p. 15; Fosberg 1955a, as Messerschmidia argentea (L. f) Johnst., which is the correct name if this plant is separated from the other species of Tournefortia of tropical America as a different genus and associated with a diminutive Siberian plant of alkaline or salty places, This question is not yet settled ) LABIATAE *COLEUS SCUTELLARIOIDES L. Coleus. Seen growing in pot in nursery in 1953, would probably not survive under Wake Island conditions without’ protection. SOLANACEAE *CAPSICUM ANNUUM L.? Pepper. A plant probably patente to this species seen growing well, flower- ing and fruiting, in protected garden plot at TAL Compound in 1953. *CAPSICUM FRUTESCENS L. Chili-Pepper. Growing well and fruiting at TAL Compound in 1952 and 1953, apparently survived the typhoon. This is the pepper with very small carrot-shaped very hot fruits. *MICOTIANA TABACUM L, Tobacco. An erect somewhat greasy-feeling herb, up to 6-8 dm. tall, with pink flowers; rare, persisting in 1952 around old Japanese garden site, status in 1952 very much as reported by Christophersen thirty years earlier. Fosberg 34448. (Christophersen 1931, p. 15; Bryan 1942, p. 213.) aiiGe *SOLANUM LYCOPERSICUM L. Tomato, Growing in raised sheltered garden plot at TAL Compound in 1952, in 1953 one plant said to have grown and fruited near the garbage dump; mentioned by Drummond-Hay (1939) as being grown in hydroponic culture. GOODENTACEAE SCAEVOLA SERICEA Vahl Scaevola; beach magnolia. ‘A low rounded shrub with bright green leaves, white flowers that appear to have been torn in half, and white fleshy fruits; occasional in scrub vegetation on Wake Islet, especially along the air strip and near Peacock Point, much less common in 1953 than in 1952, and less common in 1952 than appears from Bryan's notes in 1923 (Christophersen 1931, pp. 6-7); not seen on either Wilkes or Peale Islets in 1952 or 1953, but reported from Wilkes by Christophersen. Fosberg 33616. (Pick- ering 1876, p. 247, as Scaevola; Bryan 1926, p. 8; Christophersen 1931, p. 15; Bryan 1942, p. 213; Fosberg 1955a, p. 21, the last four as Scaevola frutescens (Mill. ) Krause. This name is based on a specimen from the West Indies which belongs to a different species. ) ‘COMPOSITAE *CONYZA BONARIENSIS (L.) Cronq. Horse Weed. A tall coarse herb with many heads of tiny whitish flowers; rare along roadsides on Wake Islet. Fosberg 34463. *LACTUCA SATIVA L, Lettuce. Seen growing in raised sheltered garden plot at TAL Compound in 1952; reported by Drummond-Hay (1939) as grown in hydroponic culture. *PLUCHEA ODORATA (L.) Cass. An aromatic grayish pubescent shrub with clusters of heads of small pale purplish flowers; common in waste places and openings, especially near the airstrip. Fosberg 33621. “XxSONCHUS OLERACEUS L, Sow-Thistle. An erect milky herb with pale yellow flowers; a rare weed around dwellings and along roadsides. Fosberg 34464, 34483. The Cryptogamic Flora Very few cryptogams are known to have been collected on Wake Island. Two brackish water green algae may be recorded, both tentatively iden- tified by Wm, Randolph Taylor. C LADOPHORACEAB RHIZOCLONIUM HOOKERTI Kg.? Forming filamentous green masses in brackish pools in bomb craters, Wake Islet. Fosberg 34479. -20- ULV ACEAE ENTEROMORPHA FLEXUOSA (Walf.) J. Ag.? Forming a dense turf on bottom of an intermittent pool, varying with tide, without water when collection was made, in bottom of an inland artificial depression, Fosberg 34475. References Cited Bailey, A. M., and Niedrach, R. J. Stepping stones across the Pacific. Museum Pict. 3: 1-63, 1951. Bryan, E. He Insects of Hawaii, Johnston Island, and Wake Island. Bishop Mus, Bull. 31: 1-94, 1926, Bryan, EZ. H, American Polynesia and the Hawaiian Chain. 1-253, Hono- lulu, 1942. Bryan, E, H. Plsnts found on Wake Island. unpub. ms. Christophersen, E, Vascular plants of Johnston and Wake Islands. Bishop Mus. Occ. Pap. 9(13): 1-20, 1931. Drummond-Hay, Lady Hay. A trip to Wake Island. Chiina Jour. 30: 333- 339, 1939. Fosberg, F. R. Northern Marshall Islands Exnedition, 1951-1952. Land Biota: Vascular plants, Atoll Res. Bull. 39: 1-22, 1955a. Fosberg F. R. Pacific forms of Lepturus R. Br. (Gramineae). Bishop Mus. Occ. Pap. 21: 285-294, 1955b. Gray, A. Botany, Phanerogamia, Part I: in U. S. Expl. Exped. 15: 1-777, 7 Pickering, C. Geographical distribution of animals and plants: in U. S. Expl. Exped. 19(2): 1-524, 1876. U. S, Dept. of Commerce, Weather Bureau. Local climatological data... 1956, Wake Island. 4 pp., Asheville, 1957. ATOLL RESEARCH BULLETIN No. 68 Additional records of phanerogams from the northern Marshall Islands by F. R. Fosberg Issued by THE PACIFIC SCIENCE BOARD National Academy of Sciences--National Research Council Washington, D, C. May 15, 1959 Additional records of phanerogams from the northern Marshall Islands by Beate Fosbergl/ The principal easily available lists of plants growing in the Mar- shall Islands north of lat. 7930' N. are in Taylor, Plants of Bikini, 1950, and Fosberg, Atoll Research Fulletin No. 39, 1955. In addition to these a list is provided in Fosberg, Military Geography of the Northern Marshalls, 1956, and there are scattered records throughout the litera- ture of systematic botany, mostly based on the collections of Chumisso, made in 1817, and referred to as from "Radack". Most of these were pro- bably collected on Odtia (the modern Wotje) and originally identified by Chamisso and Schlechtendal. The botanical names were brought up to date in the 1955 list, but further research has fesulted in a few more changes. Additional collect- ing, principally on a trip made under the auspices of the U.S. Naval Radiological Defense Iaboratory in February, 1956, to Rongelap, Rongerik, Ailinginae, Utirik, Likiep, and Kwajalein atolls has yielded a considerable number of new distributional records from these atolls. These are here placed on record, along with certain corrections in the nomenclature used in the 1955 list and in Tsylor's book. .Following the practice adopted in the 1955 list only the name of the atoll and o° the islet and collector's number are given, the numbers being in the author's consecutive series. The specimens have been or will be deposited in the U. S. National Herbarium, with duplicates, when available, in the B. P. Bishop Museum. Records are not cited from additional islets in atolls where the species concerned are already known. Brief generalized notes are added on occurrence and other matters of interest. Introduced species are indi- cated by an asterisk. Polypodium scolopendria Burm. f. Ailinginae Atoll: Sifo I,, 36675 Very local in scrub forest. *Cycas circinalis L. Likiep Atoll: ~Likiep I., 36632. Planted tree 5 m. tall, said to have been brought from Arno Atoll. Pandanus tectorius Park. Ailinginae Atoll: Sifo I, 36673. Rongerik Atoll: Eniwetak I., 36705. Rare in scrub and in Cordia forest. *Chrysopogon aciculatus (Retz.) Trin. Kwajalein Atoll: Kwajalein I., 39277. Established in lawn at air terminal, somewhat protected and semi- shaded. Y/ Botanist, U. S, Geological Survey, Washington, D. C, —2= *Cynodon dactylon (L.) Pers. Likiep Atoll: Likiep I., 36726. In lawn around church, “Digitaria ciliaris (Retz.) Koel. — Kwajalein Atoll: Kwajalein I., 39499. Occasional in waste ground near airstrip; probably a recent intro- duction. ‘Digitaria insularis (L.) Henr. Listed in 1955 as Tricachne insularis (L.) Nees, but referred to as Digitaria by Henrard in his monograph of Digitaria. Digitaria pruriens var. microbachne (Presl) Fosb. Ailinginae Atoll: Sifo I., 36688. Found only in one small opening in the forest. Listed in 1955 as Digitaria microbachne (Presl) Henr. but better regarded as a variety of the widespread D. pruriens. ostis See ae oe oe ee "aonatatk Atoll: zouk. ae i 36699. Common in cleared strip. lepturus gasparricensis Fosb, “This was reported from Pokak Atoll in the 1955 list as_Lepturus sp. It has the distinction of being one of the few species of vascular plants endemic to coral atolls, being known only from Pokak and Wake Atolls, Lepturus repens (Forst. f.) R. Br. Since the 1955 list a discussion of the Pacific forms of Lepturus has been published (Fosberg, Occ. Pap. Bishop Mus. 21: 285-294, 1955) and a number of varieties were recognized in Lepturus repens. In the northern Marshall Islands the following were found to occur, of which most of the specimens were cited in the 1955 list under L. repens: — Var. repens from Eniwetok Atoll, var. subulatus Fosb. from Eniwetok, Bikini, Utirik, Ujelang and Wotho atolls, var. septentrionalis Fosb,. from Pokak, Bikar, Eniwetok, Bikini, Rongerik, Taka, Ailuk, Jemo, Likiep, Kwajalein, Lie, and Ujelang atolls, and var. occidentalis Fosb. from Eniwetok Atoll. To these may be added the following new records: var. subulatus, Kwajalein Atoll, Kwajalein I., 36642, and var. septentrionalis, Ailinginae Atoll, Sifo: 1. . ” 36673; Rongelap Atoll, Rongelap I., 36652, Eniaetok I., 36664, and Kzbelle I., 36667. *Paspalum finbriatum HBK Kwajalein Atoll: Kwajalein I., 36644. Established in waste ground near the airstrip, previously known in the Pacific from Hawaii and Guam, *Paspalum setaceum Michx. Kwajalein Atoll: Kwajalein I., 36641 (determined by Aunes Chase), 39488. : Well established in waste ground near the airstrip, not previously reported from the Pacific Islands, native of southeastern United States. aie Thuarea involuta (Forst.) R. & S. “ilinginae Atoll: Sifo I., 36696. Forming ground cover in some areas in open forest. *Cyperus polystachyos Rottb., Kwajalein Atoll: Kwajalein I., 39490. Local along airstrip. *Cocos nucifera L. Ailinginae Atoll: Sifo I., not collected, Rongerik Atoll: Eniwetak I., not collected. Several small groves on Sifo Islet in the interior. *Rhoeo spathacea (Sw.) Stearn - Usually known as R, discolor (L'Her.) Hance and thus reported in 1955, but R. spathacea is an earlier name and the correct one for this widely planted tropical ornamental (see Stearn, Baileya 5: 195-198, 1957). *Crinum asiaticum L. Planted generally on Kwajalein Islet as an ornamental. *Crinum bakeri Engl. UtinekiAnolils OUtirik Ls, 136713: Planted along paths in village. This species has hitherto been known only from the original collection, made by Finsch on Mille Atoll, southern Marshalls in 1888. It was previously regarded as possibly based only on a badly prepared or abnormal specimen of C. asiaticum, in which the perianth tube is shorter than the lobes, However, the Utirik plants, seemingly perfectly normal, have this character, as well as crispate margined perianth segments. *Crinum latifolium var. zeylanicum (L.) Hook, f. ex Trim. Likiep Atoll: Likiep I., 36723. Planted in gardens. *Hippeastrum puniceum (Lam. ) Urban Likiep Atoll: Likiep I., 36727. Planted in gardens. Tecca leontopetaloides (L.) 0. Ktze. Ailinginae Atoll: Sifo I., 36692. Common in forest and coconut groves. *Casuarina equisetifolia L. Likiep Atoll: Likiep I., 36630. Tree 8m. tall growing in garden on Likiep; planted also on Kwaja- lein and there apparently occasionally spontaneous from seeds of planted trees. *Peperomia obtusifolia (L.) A. Dietr. Likiep Atoll: Likiep I., 36724. Growing in pot in shelter of buildings. Le Peperomia ponapensis C. DC. ( Recorded from Lae Atoll as Peperomia sp. in 1955, specimen since identified by T. G. Yuncker. Fleurya ruderalis (Forst. f.) Gaud. ex Wedd. Rongelap Atoll: Gegen I., 36737; Kubelle I. 36671; Ailinginae Atoll: “Sifo I., 36691. Locally common, esnecially on rocky ground, plants on Gegen in very poor condition. *Artocarpus altilis (Park.) Fosb. Rongelap Atoll: Rongelap I., planted in village, fruit preserved for analysis, but no herbarium specimen made, *Artocarpus mariannensis Trec. The following records, now referred to A, mariannensis were in 1955 reported as A. altilis: Utirik Atoll, 33691. Ujae Atoll, 34417. Iae Atoll, 34091. Ailuk Atoll, 33924, 33057 Likiep Atoll, 33840, 36633. In Baceenl a specimen from Eniwetok Atoll, St. John 23714 and one from Rongelap Atoll, Taylor 46 1399; both in the U.S. National Herbarium seen to belong to A. mariannensis., From Wotho Atoll, Fosberg 34233 and from Bikini Atoll, "Taylor "46-1146 seem to be hybrids between A. altilis and this species, Boerhavia tetrandra Forst. f. Ailinginae Atoll: Sifo I., 36685. Common on sand and cobble flats. Boerhavia sp. Rongelap Atoll: Kabelle I., 36670; Gegen I., 36728. Rongerik Atoll: Enivetak I., 36700. Common to dominant in ground cover in open and thin forest. Tris is a white-flowered form with leaves resembling those of B,_mutabilis R, Br., and its exact status is uncertain as yet. It is also known from Wake and Johnston Islands. It would ordinarily be referred to B, diffusa L. and may be what Taylor (1951) has reported under that name from the Marshall Islands, It was mentioned in the 1955 list from Wake. *Bougainvillea glabra Choisy Utirik Atoll: Utirik I., 36716. Likiep Atoll: Likiep I., 36629, 36631. . ‘ Several color forms planted around houses but not growing es- pecially well, Pisonia grandis R. Br. bk. ast Ailinginae Atoll: Sifo I., 36680. Codominant with Cordia in the forest, as in other dry atolls in the northern Marshalls and Wake Island. Achyranthes canescens R, Br. Rongelap Atoll: Gegen I., 36730. Rare in thin Pisonia forest. a5 Portulaca lutea Sol. Rongerik Atoll: Eniwetak I., seen by Evans, but not collected, Ailinginae Atoll: Sifo I., 36676. Occasional in sandy openings. Portulaca samoensis v. Poelln. Rongelap Atoll: Rongelap I., 36658. Very local in sandy soil in grassy opening in coconut plantation. Cassytha filiformis L. Ailinginae Atoll: Sifo I., 36689. Occasional, parasitic on Guettarda, *Delonix fegia (Bojer) Raf. Likiep Atoll: Likiep I., 36638. Old tree, planted in garden. *Leucaena glauca (L.) Benth. Likiep Atoll: Likiep I., 36636. Planted in garden. *Mimosa pudica L. Kwajalein Atoll: Kwajalein I., 36640. Very local on disturbed ground near air-strip; plants very chlorotic. Soulamea amara L. Rongerik Atoll: Enivetak I., 36706. Small colony in Pisonia forest; new northernmost record in Marshall Islands, Suriana maritima L, Ailinginae Atoll: Sifo I., 36690. Common locally in scrub. *Euphorbia cyathophora Murr. The plants reported in the 1955 list as Euphorbia heterophylla L. have been redetermined by Dr. R. L. Dressler as E. cyathophora. *Euphorbia glomerifera (Millsp.) Wheeler Reported in 1955 list as Ey,horbia hypericifolia L. but probably better referred to E, glomerifera. Euphorbia hirta L. Berrie Witold: Utirik f.,’ 3ovdae Several plants seen near landing. *Pedilanthus tithymaloides (L.) Poit. The variegated form of this is planted around houses on Kivajalein Iset. *Phyllanthus amarus Schum. & Thonn. Eniwetok Atoll: Hmniwetok I., St._John 23,706. (US) Reported in 1955 list as Phyllanthus niruri L., but that species has been shown by Dr. G. L. Webster not to occur in the Pacific. The plants so referred from the Marshall Islands have been redetermined by Webster to be P. amarus. ies Allophylus timorensis Bl. Cee Ailinginae Atoll: Sifo I., 36673. Single clump in forest in center of islet. Triumfetta procumbens Forst. Ailinginae Atoll: Sifo I., 36632. Common in open forest and openings. Sida fallax Walp. Rongerik Atoll: Eniwetak I., 36698. Very local in scrub forest on back slope of lagoon beach ridge. *Cucurbita maxima L,? Rongelap Atoll: Rongelap I., 36655. Persisting from cultivation around abandoned village. Terminalia samoensis Rech, Ailinginae Atoll: Sifo I., 36683. Rare in scrub. *Centella asiatica (L.) Urb. Utirik Atoll: Ytirak) 1, , 36714. Several patches seen in coconut plantation. *Inomoea triloba L. Kwajalein Atoll: Kwajalein I., 39487. Very local, one patch seen in waste ground near airstrip. Ipomoea tuba (Schlecht. ) G. Don Ailinginae Atoll: Sifo I., 36693. Common in openings. Cordia subcordata Lam. Ailinginae Atoll: Sifo I., 36695. Co-dominant in forest with Pisonia grandis. Tournefortia argentea L. f. Ailinginae Atoll: Sifo I., 36679. Very common tree in scrub forest. Reported in 1955 list as Messerschmidia argentea, but should probably be replaced in Tournefortia. Clerodendrum inerme (L.) Gaertn. Ailinginae Atoll: Sifo I., 36697. Common in small openings in forest. “Jitex trifolia L. Kwajalein Atoll: Kwajalein I., 36709. Planted as an ornamental around buildings. *Russelia equisetiformis Schlecht. & Cham. UtirikAtToMey) Utirik | 1., 36722. Planted around an outbuilding. —T *Pseuderanthemum carruthersii var. atropurpureum (Bull) Fosb. Utirik Atoll: Utirik I., 36721. Planted as an ornamental. A very common ornamental throughout the Marshalls, reported in the 1955 list as Pseuderanthemum atropurpureum,. It should at most be regarded as a variety of P. carruthersii Seem., as the most obvious differences are in the color of leaves and flowers. Guettarda speciosa L. Ailinginae Atoll: Sifo I., 36631. Common generally. *Morinda citrifolia L. Ailinginae Atoll: Sifo I., 36687. Common around coconut groves, not seen elsewhere on islet. Scaevola sericea Vahl Ailinginae Atoll: Sifo I., 36677. Dominant in scrub at top of beach and to some extent inland. Recorded in 1955 list as Scaevola frutescens (Mill.) Krause, but that name is not correctly applied to the Pacific Island plant. *Bidens pilosa var. radiata Sch.-Bip. Kwajalein Atoll: Kwajalein I., 36639. Abundant locally in waste places near airstrip, probably introduced from Guam, where it has appeared recently and spread with great rapidity. Wedelia biflora (L.) DC. Ailinginae Atoll: Sifo I., 36694. Common in openings and edges of forest. In addition to the foregoing, which are either indigenous to the Marshalls, naturalized, or growing under normal cultivation, a large number of cultivated species have been brought to Kwajalein and planted in pots in a nursery, and in a specially prepared well-protected garden. These establishments were well-cared for and most of the plants were flourishing in 1956. They had been fertilized with the blue-green alga, Nostoc, swept up from the water catchment, and, when they seemed to need it, very lightly with commercial fertilizer, Fresh vegetables of a num- ber of sorts were being produced in the garden, a feat that has been accomplished elsewhere on coral atolls only with hydroponic techniques. In the previous year over a ton of Chinese cabbage had been produced, according to U.S. Navy Chief R.K.W. Lee, the man with a "green thumb" who was responsible for this enterprise. He had also started to plant out, in the shelter of the buildings and installations on the base a few of the ornamentals that seemed to be doing best under the nursery condi- tions. In 1958 the garden had been discontinued and the plants in the nursery, in the shelter of the old air terminal building, seemed to be largely caring for themselves. Most of them were still in fairly good condition and had made considerable growth. A list of the species observed, arranged alphabetically, is added here, as an indication of what species can be cultivated under atoll conditions if given pro- tection and fertilization. Judging by the fact that Nostoc seemed to be of value as a fertilizer, nitrogen deficiency may be one reason why some plants do not thrive on atolls. A number of other species were growing in the nursery, but were unfamiliar and not identifiable from the mate- rial seény It is probable that a few of the species listed may eventually prove adapted to general cultivation in the Marshalls, but in all pro- bability without special attention few of them will become established, It is reported that a selection of the nursery stock from Kwajalein has been taken to Eniwetok for planting there, but no information is yet available as to whether any of the plants have been successful. Acalypha hispida Acalynha wilkesiana Aleurites moluccana Allamanda cathartiéa var. hendersonii Allium fistulosum Alocasia macrorrhiza Alpinie purpurata Ananas comosus Angelonia angustifolia Anthurium sp. Apium graveolens Apium petroselinum Araucaria excelsa Argyreia sp. Artocarpus altilis Asparagus sprengeri Asplenium nidus Asystasia gangetica Barleria cristata Barringtonia asiatica Bauhinia sp. Begonia sp. Beloperone guttata Bougainvillea spectabilis Brassaia actinophylla Brassica napus Brassica oleracea Brassica pekinensis? Breynia nivosa Caladium bicolor Calophylium inophyllum Canna indica Capsicum annum Carica papaya Cassia alata Casuarina equisetifolia Catharanthus roseus Cerbera sp. Chrysolidicarpus lutescens Cibotium sp. Citrullus vulgaris Clerodendrum inerme Coccoloba uvifera Cocos nucifera Codiaeum variegatum Coffea arabica Coleus scutellaria Cordia sebastena Cordyline australis Cordyline terminal is Cordyline sp. Crinum asiaticum Cryptanthus zonatus Cryptostegia grandiflora Cucumis melo var. Cucumis sativus Cucurbita pepo Cucurbita sp. Cymbidium sp. Cycas circinalis Daucus carota Delonix regia Dieffenbachia sp. Duranta repens Epidendrum sp. Eoipremnum pinnatun Erythrina variegata var. orientalis Euphorbia cyathophora Euphorbia nereifolia Euphorbia pulcherrima Ficus rubiginosa Ficus microphylla Gardenia jasminoides Gliricidia sepium Gomphrena globosa Guettarda speciosa Hedychium coronarium Hemigraphis sp, '.:.* Hernandia sonora Hibiscus mutabilis Hibiscus rosa-sinensis Hibiscus schizopetalus Hibiscus tiliaceus Hibiscus (ornamental hybrid) Hymenocallis littoralis Ipomoea batatas Ixora casei Jasminum pubescens Kalanchoe pinnata lactuca sativa Lagerstroemia indica Macadamia ternifolia Merremia tuberosa Mirabilis jalapa Morinda citrifolia Murraya paniculata Musa sapientum Nephrolepis hirsutula Nerium oleander Nerium sp. Ochrosia ovpositifolia Osteomeles anthyllidifolia Pandanus tectorius Passiflora edulis Pedilanthus tithymaloides Pemphis acidula Pentas lanceolata Persea americana Petraea volubilis Philodendrum oxycardium Phoenix sp. Phyllodendron sp. Pilea microphylla Pisonia grandis Plumeria obtusa Plumeria rubra Podocarpus sp. Polypodium scolopendria Polyscias fruticosa Polyscias guilfoylei Polyscias scutellaria Polyscias tricochleata Polyscias sp. Pritchardia sp. owe Scindapsus aureus Sesuvium portulacastrum Solandra sp. Solanum lycopersicum Strelitzia reginae Tabebuia pentaphylla Tabernaemontana coronata Theophrastia sp. Thespesia populnea Thevetia peruviana Vanda "Miss Joachim" Vitex trifolia Zea mays Zinnia elegans Pseuderanthemum carruthersii var. atropurpureum Pseuderanthemum carruthersii var. carruthersii Pseuderanthemum carruthersii (var. with yellow veins) Ptychosperma macarthuri Raphanus sativus Reavenala madagascariehsis Rhapis flabelliformis Rhoeo spathacea Ruellia longifolia Russelia equisetiformis Sansevieria roxburghii Scaevola sericea Schinus terebinthifolius 1 ATOLL RESEARCH BULLETIN No. 69 Contribution to a German Reef-Terminology by Georg Scheer Issued by THE PACIFIC SCIENCE BOARD National Academy of Sciences--National Research Council Washington, D. C. May 15, 1959 WW Contribution to a German Reef-Terminology by Georg Scheer!/ In Germany recent publications on coral reefs are lacking due to isolation before, during, and after the war. The younger generation of zoologists has had no opportunity to undertake expeditions to tropical seas. But the interest in coral reefs and their inhabitants has always been keen in Germany. Dr. Hans Hass, the founder of the International Institute for Sub- marine Research, Vaduz (Liechtenstein), has placed his research ship, the yacht "Xarifa' at the disposal of science. In 1953-54 he carried out an expedition to the Caribbean Sea and to the Galapagos Islands in the Eastern Pacific. He led another expedition in 1957-58 to the Red Sea and to the Maldive Islands in the Indian Ocean. I am much obliged to Dr. Hass for giving me the opportunity to take part in both expeditions. During both voyages we were diving with oxygen and compressed-air diving equipment. In this way we were able to study the coral reefs in their many aspects by direct observation. We worked in a team, one col- lecting fishes, others making behavioural studies and so on. Dr. Hass and I were particularly interested in the formation of coral reefs and in the sociological composition of their different parts. Soon we realized the necessity to use a clearly defined nomencla- ture for the different parts of a coral reef, and to coordinate this German nomenclature with the English one. After long discussions with the other members of expedition (Dr. Hans Hass; Prof. Dr. W. Luther, director of the Zoological Institute of the University Darmstadt; Dr. I. Eibl-Eibesfeldt, Max Planck-Institute for Verhaltensphysiologie, Seewiesen; Dr. L. Franzisket, director of the Museum of Natural History, Muenster; Dr. S. Gerlach, Zoological Institute of the University Kiel; and Dr. W. Klausewitz, Research Institute and Museum Senckenberg, Frank- furt a.M.) may I present in the following lines a few German terms with English equivalents. They concern mainly atolls, but they are applicable in like manner to all other reefs. This terminology does not go for the present into details, but handles only the great striking characteristics of reefs. The following list contains mostly translations of English expres- sions, because there are not such far-reaching subdivisions of coral reefs in the older German pub:..ications. I take as basis for the termin- ology the publications of Tracey, Uioud and Emery: "Conspicuous Features of Organic Reefs"2/, of Harry: “Ichthyological Field Data of Raroia Atoll, Tuamotu Archipelago" 3/ and of Kuenen: "Classification of Reefs and their Component. Parts"4/. All these publications include figures il- lustrating the English names mentioned in this paper. 1. Zoological Department, Hessisches Landesmuseum, Darmstadt, Germany. 2. Atoll Research Bulletin No. 46, 1955 3. Atoll Research Bulletin No. 18, 1953, p. 21-26 4, Marine Geology, New York 1950, chapter 6: Coral Reefs, p. 423-430 I on REEFS IN GENERAL fringing reef barrier reef atoll reef grown up independently in lagoons of barriers and atolls drowned reef uplifted reef SEAWARD REEF outer slope terrace bench or shelf under-reef cavern reef front surge channel surge zone surf zone buttress groove and buttress-system reef margin algal ridge RIFFE, ALLGEMEIN Kuestenriff or Saumriff Wallriff or Barriereriff Atoll Fleckenriff or Flachseeriff ertrunkenes Riff gehobenes Riff AUSSENRIFF Aussenabhang, if the slope is not too steep; Aussenwand, if the slope is steep, about 70 to 90° to level Terrasse, a broad, little inclined step in the slope Absatz, a small terrace, interrupt- ing the slope Hoehle, great cavern; Grotte, small cavern Riffabhang; Riffwand, the strongly inclined part of the slope, which is more or less covered with living corals Brandungsrinne Wellenzone, the outward part of a surge channel Brecherzone, the inner, landward part of a surge channel (Wellenzone and Brecherzone together are the Brandungszone ) Riegel Rinnensysten Riffkante Lithothamnionwall REEF FLAT seaward edge tidepool a sandy zone in shallow water in front of the coast inter-island channel ISLAND seaward beach shingle rampart boulder rampart beachrock lagoon beach LAGOON REEF FLAT a sandy zone in shallow water between the island and the lagoon lagoon reef margin lagoon reef face LAGOON lagoon slope lagoon floor coral head, knoll, or mound Acropora thicket tall narrow pinnacle patch reef miniature atoll -3- Riffkanal Strandsandstein RIFFPLATTE Aussenkante ! Fluttuempel seeseitige Seichtwasser-Sandzone INSEL Aussenstrand or Meeresstrand Geroell-Strandwall ! Block-Strandwall | Lagunenstr and LAGUNENRIFFPLATTE lagunenseitige Seichtwasser- Sandzone Innenriffkante Innenriffhang LAGUNE ' Innenhang Lagunenboden | Korallenblock / Acroporenhecke Turm Buckelriff or Korallenbuckel Miniaturatoll, an atoll-shaped patch reef with a ring of living corals around a deeper flat cov- ered with sand and dead, broken corals ahs faro Faro, an atoll-shaped reef of greater extension with a dis- tinct lagoon This list may serve as a base for a more elaborated German reef terminology. With greatest interest we are looking forward to the Reef Terminology Index, which the Pacific Science Board of the National Academy of Sciences-National Research Council, Washington, D. C., is pre- paring. ATOLL RESEARCH BULLETIN No. 70 Atoll News and Comment Issued by THE PACIFIC SCIENCE BOARD National Academy of Sciences--—National Research Council Washington, D. C. May 215, 1959 ATOLL NEWS AND COMMENT With this issue is started a series of informal articles by the editors of the Atoll Research Bulletin calling attention to expeditions to atolls and reefs, to record notes and observations that are too short to appear as separate numbers of the Bulletin, to call attention to and briefly review significant new publications, and to point out and empha- size important items from older publications pertinent to the modern study of coral atolls. The editors will welcome news notes, observations, and other items of interest that readers may care to send in, and will give the senders proper credit for any such used. In the past the editors have gone to considerable trouble to ascer- tain correct addresses for people on the Bulletin Mailing list who have moved or are reputed to have moved. We have decided that this is a fruitless waste of time. Hereafter, when a copy is returned the addressee's name will be removed from the list without further ado and will only be restored if he writes and requests it, This may make it possible to fill some of the many requests that must now be turned down. ARB readers may be interested in the progress of the Pacific Science Board's Reef Terminology Index project. Since the Tuird Cir- cular of the project, with a selected bibliography of the geological literature on reefs almost half of the papers on reef geology have been examined and about 1400 definitions and usages have been abstracted. The remaining geological papers have been assigned to members of the Inter- national Committee for examination. A bibliography of the principal Papers on reef ecology, botany, and zoology is in preparation. The aim of this program is to bring together the terms in English applied to reef features, their original definitions and differing later usages, foreign language equivalents, principal genera of organisms found in and on reefs and atolls, locations and brief characterizations of classical reefs, and any other information that will possibly lead to clearing up the mess in which terminology referring to reefs and atolls finds itself. Cooperation on this project is earnestly invited, Corres- pondence may be addressed to the ARB editors. We are glad to bring to the attention of the readers of the Atoll Research Bulletin the International Institute for Subacrine Research, Vaduz, Liechtenstein, with its floating biological laboratory, the research vessel "Xarifa", This vessel is equipped to carry eight scien- tists and is fully self-contained and capable of carrying out many types of marine biological research as well as other types of scientific work on islands and coast-lines that are hard to reach by ordinary means. In 1958, work was carried out in the Maldives, and future programmes call for work out of northern Australia into the Molucea, Phili pine, and Caroline Islands. Cooperation is invited from other interested insti- tutions in expeditions on a semi-annual and annual basis. The presi- dent of the Institute is Dr. Hans Hass, the scientific director, Dr. Trendus Eibl-Eibesfeldt. They are supported by a Committee for the Exploration of Tropical Seas and Islands, made up of internationally known scientists from the fields of Zoology, Anthropology, Ethnology, Geophysics, Paleontology, Geology, and Botany. ieee The Pacific Naturalist, published by the new Beaudette Foundation for Biological Research, Box 227, R.F.D. 1, Solvang, California, U.S.A., is the latest addition to the assortment of journals in which the re-~ sults of research on coral atolls may appear. Tuais serial, of which five numbers have appeared since its initiation, Dec. 17, 1958, is handsomely printed, beautifully illustrated, and will appear irregularly as a medium for publication of the results of the work of the Beaudette Foundation. ' We take occasion to welcome this new institution into the Pacific science community, and hope that its "program of inquiry into the systematics, distribution and utilization of marine and littoral organisms, particu- larly the plants, of the eastern tropical and subtropical Pacific" may prosper, The first five papers, including the one reviewed below are a good start. The president of the Beaudette Foundation is Palmer T,. Beaudette and its scientific director is our friend E, Yale Dawson. Mr, Bernd Lambert, of Sydney, Australia, writes that he will leave shortly for approximately 1% months of anthropological field work in the northern Gilbert Islands under a grant from the Tri-Institutional Paci- fic Program. Recent Expeditions to Atolls ~ While it is probable that the resident scientist, in any region, may eventually achieve deeper understanding of his environment than does the expedition scientist, the geographical facts of life seem to indicate that most scientific work on coral atolls will, for some time to come, be done by expeditions. The very limitation of time inherent in expedi- tion work may well have its beneficial aspects. The knowledge that time is limited, the novelty of the situation, and, often, the company of the members of the expedition may all have the effect of stimulating the expedition scientist, both mentally and physically, and on a successful expedition a prodigious amount of work may be accomplished. Most readers of the Atoll Research Bulletin are either past or potential members of expeditions to coral atolls. They may be interested in brief notes about recent expeditions to atolls, the results of which are not yet available in published forn. Maldive Islands: Dr. Georg Scheer writes that the Xarifa (see above) spent some months in the Indian Ocean, concentrating in this group of atolls, under the direction of Dr. Hibl-Eibesfeldt. large collections of marine animals and birds were made and direct observa- tion was carried out by means of under-water equipment. Jaluit Atoll: After the visit of disastrous typhoon Ophelia, a Pacific Science Board expedition, under ONR sponsorship, spent a week on the atoll to study the effects of this typhoon. The party, under the leadership of David I. Blumenstock, consisting of A, H, Banner, F, R. Fosberg, J. L. Gressitt, E. D. McKee, J. B. MacKenzie, and H, J. Wiens, collected substantial amounts of data on what the storm did to the physiography, soils, fauna, flora, vegetation, and human life on the atoll, This will make a substantial contribution toward understanding the role of typhoons in the ecology and geomorphology of atolls in gen- eral. For a first preliminary announcement of results see Nuture 182: 1267-1269, 1953. ae Tuamotus: Dr. Jacques Barrau, Plant Introduction Officer of the South Pacific Commission, recently spent some weeks in the Tuamotus and is currently in Micronesia, where he has visited several atolls among the other islands. He is actively collecting and studying the forms of breadfruit in an attempt to understand the origin and migration of the many horticultural varieties. Important light may be shed on the pre- historic migrations of the Pacific peoples, in addition to solving the obvious botanical problems. Clipperton Island: As an adjunct to the DOLDRUMS Expedition of the Scripps Institution of Oceanography last summer a party of biolo- gists spent several weeks making a scientific survey on Clipperton Island, a ring-shaped coral island, like an atoll except for a snag of volcanic rock sticking up on one side, This island is of great interest in that it is one of the most isolated of all coral atolls and the only one in the eastern Pacific. Its biota is extremely impoverished. The ARB assistant editor was enabled to serve as botanist and geologist of the expedition through the kind invitation of Scripps Institution and financial assistance from the Joseph Henry Fund of the National Academy of Science. Some of the results may be expected in an early issue of ARB, Canton Island: Under the auspices of the Civil Aeronautics Adminis- tration (now the Federal Aviation Arency) Dr. and Mrs. Otto Degener revisited Cnton Island in early 1958 to inspect the results of attempts at revegetation carried out in 1950 and 1951 (see ARB 41-43). Part of the results are reported elsewhere in this issue. Rongelap Atoll: During late February and early March, 1959, a Brookhaven National kaboratory party under Dr. Robert A. Conard visited Rongelap Atoll for the yearly examination of the Marshallese people who were affected by radioactive fallout in 1954. Dr. Baruch S, Blumberg, a member of the party, during his free time made valuable observations for the Pacific Ve.etation Project in connection with the possible Jong- term effects of fallout on plants. The notes collected will be made available in this issue or the next of ARB, While on Rongelap the party enjoyed the company and cooperation of a party from the University of Washington Applied Fisheries Laboratory, under Mr. Edward Held, carrying out their semiannual resurvey of radioactivity under the AEC program. This has now evidently expanded to include serious work on the land en- vironment, with investigations of soil radioactivity. Dr, Frunk Richardson was studying the birds of the atoll. Current Literature Geology of Kepingamarangi Atoll, Cxroline Islands, by Edwin D. McKee, Bull. Geol. Soc. Amer. 69: 241-278, 1958; and Sedimentary Belts ih lagoon of Kapingamarangi Atoll, by Edwin D. McKee, John Chronic, and Estella B, Leopold, Bull. Amer. Assoc. Pet. Geol. 43: 501-562, 1959. It is a pleasure to be able to report two excellent papers embodying the final results of the geological and soils investigations made on the Pacific Science Board expedition to Kevingamarangi, 1954. In addition to what might be expected in geological reports, much infor- mation of an ecological nature, soil chemistry, a diatom flora, reports of pollen and dinoflagellates, and of recent foraminifera are included. Comparisons are made with geologically ancient reefs. Mh ey je A Review of Ciguatera, Tropical Fish Poisoning, with a Tentative Explanation of its Cause, by J, E. Rendall, Bull. Marine Sci. Gulf and Csrib, 8: 236-267, 1958, will be of great interest to atoll students, as poisoning from eating reef fishes is a widespread hazard on atolls, Jack Rendall has advanced what seems to be the first reasonable suggestion as to the cause of this poisoning. He suggests that the fish may acquire the toxin from eating blue-green algae that characterize the pioneer vegetation on newly denuded substrata. This would account for the seemingly inexplicable inconsistency in the toxicity of the same species of fish at different times and places, Changes in Palmyra Atoll and its Ve,etation through the Activities of Man, 1913-1958, by E. Yale Dewson, Pacific Naturalist, 1(2): 1-51, 1959, is a most interesting paper, handsomely presented. The destruc- tive character of the influence of man is as appalling on Palmyra as in most other places, Interesting is the fact that interference with the circulation of sea water in and out of the lagoon resulted in complete destruction of the lagoon biota. The quick recovery of the land vege- tation is a phenomenon also observed elsewhere on atolls. The complete lists of land and marine plants collected are valuable. Dawson also makes the suggestion that the pioneer blue-green algae, especially lyngbya, have something to do with the sudden appearance in 1943 of Ciguatera or poisoning from eating reef fishes (see Randall article noted above). Dating the latest movements of the Quaternary sea level, by Rhodes W, Fairbridge, Trans. N. Y. Acad. Sci. II, 20: 471-482, 1958, furnishes a welcome review of recent data on eustatic sea level fluc- tuations. An understanding of these fluctuations and their chronology and amplitude is a necessity to one who would think broadly on the theories of the origins of atolls, on their surface geology, or on their land ecology. Data on recent attempts to date some of the fluctuations by radiocarbon methods makes it possible to have a bit more confidence in actual figures on the postglacial lowerings of sea level that pro- vided most of the land surface on atolls. Flora of the Hall Islands, by Benjamin Stone, Pac. Sci. 13: 88- 104, 1959, gives brief descriptions of the vegetation and a list of the known species of vascular plants of this group of atolls in the northern part of the Carolines. This paper will help materially in filling out the gaps in our knowledge of the Curoline atolls, mostly to date very poorly known. — The Ecology of Jamaican coral reefs I, Species composition and zonation, by Thomas F, Goreau, Ecology 40: 67-90, 1959, is comprehen- sive, well illustrated, and will contribute significantly to an under- standing of the coral reefs and islands in the West Indies. The Elepaio, vol. 19, no. 9, has two articles on atoll birds, an account of blue-gray noddies and of wnite throated storm petrels on Christmas Island, by M. D. Gallagher, and one of the old-squaw duck on Midway by E. H. Bryan. It is of interest that the nesting habits of the blue-gray noddy (Procelsterna) on Christmas do not match exactly those seen on Bikar and Pokak Atolls in the Marshalls. ae It may be appropriate here to call attention to the many valusble articles on atoll birds and atoll ecology scattered through the 19 vol- umes of Elepaio. This publication of the Hawaiian Audubon Society must be regarded as an important source of information on atolls, esrecially the Hawaiian ones. Coral Island, by Marston Bates and Donald P, Abbott, 1-254, Seribners, N.Y., 1958, is an example of popular science writing at its best. The authors give an informal account of the Pacific Science Board expedition to Iivaluk, a portrait of an atoll, an autobiographical account of scientists at work, and a liberal introduction to scientific natural history, in a book as easy to read as a light novel, Highly recommended reading is the chapter on The Reefs, Here is much of the essence of Marine ecology. Here, also, is reassurance to the reviewer as to the adequacy of his education as a biologist--he has had the two requisites for its completeness, opportunities to study both coral reefs and tropi- cal rain forests, Not the least of the virtues of this volume is the sharing with the reader of the authors! scientific philosophies, Marine Ecology, by H. B. Moore 1-493, Wiley, 1958, is a compre- hensive book, reviewing much of the important work that has been done in the field of marine ecology and boiling it down to the point where it is possible to assess the field, It is essentially a review, and throughout the text the sources of the material summarized are faith- fully cited. For this very reason the book is really not very well adapted for use as a text book except for very advanced courses, It is not smooth or easy reading. Rather it is a compendium of informa- tion so arranged that what is available on almost any type of marine environment and its inhabitants may be readily located. The coral reef environment is treated at somewhat less length than that allotted to some of the others, but the discussion is an important summary, emphasizing depth, light, and temperature limita- tions. Of especial interest is the attention given the zooxanthellae, or intracellular algae found in reef corals and other organisms. The illustrations of reef communities are especially striking, mostly taken from Manton and Stephenson's work on the Great Barrier Reef. Reefs such as those found in the Central Pacific that are largely algal in ‘nature, receive scant attention. The ecological framework around which the book is built is that of the marine environment, its subdivisions and its gradients. Eco- logical behavior, life histories, and dif“ering requirements of larval and adult stages are properly emnhasized. Analysis of data is done very thoroughly, synthesis scarcely attempted. Curiously, the term ecosystem is completely absent, nor is any equivalent used or proposed. Intracommunity relationships are discussed at length throughout the book, and dynamics are stressed here and there. The viewpoint, however, seems to be basically autecological, This seems true, at least if one regards the cited and quoted material as indicative, in spite of the fact that the author in his introduction emphasizes the necessity for integration (of the results of autecology and symecology). He even coins the term "integral ecology" for a third subdivision which he regards as coordinate with autecology and synecology. Perhaps some of his readers may fail to see the difference between "integral ecology and synecology". LL Wb. es — ee One may also wonder if the author's question whether the autecolo- gist's study of individuals can ever adequately describe communities as they occur under natural conditions, on p. 419, means that he thinks the ultimate objective of autecology is to describe communities. He also, on the same page, betrays some vagueness on the matter of what indicates that the community is more than the sum of its parts. No men- tion is made of effects caused by the community that could not con- ceivably result from its members separated in space or not organized in. the way that the community is organized. Organization of communities is scarcely stressed at all in the book. Land ecologists will wonder if the lack of emphasis on climax and his statement on p. 420 that at no time are we dealing with a really stable climax condition is an indica tion of the author's bias or if such stability in communities as is observed on land simply does not exist in the sea, Moore, in this book, has presented a magnificent accomplishment in bringing such a vast array of data into such order that it may be examined. Reef specialists will be disappointed in the scant attention given their part of the marine field. However, all of them will find the book extremely useful and interesting. Not the least useful feature will be the 27 pages of bibliography and the classified list of the genera referred to in the text. It is a pity that as much cannot be said for the editing as for the compilation itself. Typographical errors abound, including some in the bibliography, For example Anton Bruun's name is spelled Braun on p. 211 and Bruan on p. 431. The most annoying thing of all, however, is the frequent reference to organisms by binomials with the generic name re- duced to its initial. This is entirely proper if the genus has been spelled out earlier on the page, but in literally dozens or perhaps hundreds of cases one looks tack a long way for the full binomial. For example, L, littorea on pp. 22 and 36 refers back to Littorina to on pp. 21 and 22, In general far more familiarity with individual organisms is assumed than is warranted. This is partly offset by the list of genera of organisms referred to arrangec according to the taxo- nomic system on pp. 421-428, with cross references from the index. Men- tion of this cross referencing scheme in the preface would have saved valuable time spent in discovering it. A glossary would have helped, especially where such words as "systrophe" (p. 153) and "implode" (p. 251) are used. However, far less unfamiliar words and usages are found here than is usual in ecological works, Older Literature It is interesting to note, on rereading Charles Darwin's des- cription of Keeling Island in his Journal of Researches (London, 1839, pp. 539-553), that in addition to his many other great discoveries and generalizations he pointed out the essential principle of what has since been called the Ghyben-Herzb?:s fresh-water lens and that this wa. first noticed on an atoll islet. He says (np. 545-546), "On this island the wells are situated from which ships obtain water, At first sight it appears not a little remarkable that the fresh water should eae regularly ebb and flow with the usual tide. We must believe that the compressed sand or porous coral rock acts like a sponge; and that the rain water which falls on the ground being specifically lighter than the salt merely floats on its surface, and is subject to the same movements. There can be no actual attraction between salt and fresh water, and the spongy texture must tend to prevent all admixture from slight distur- bances. On the other hand, where the foundation consists only of loose fragments, upon a well being dug, salt or brackish water enters; of which fact we saw an instance on this same island." Additional note A specimen of scarab collected on Male Atoll in the Maldive Islands in 1956 by F. R. Fosberg has just been identified by Dr. Oscar Cartwright, of the Division of Insects of the U. S, National Museum as Oryctes rhino- ceros (L.), the rhinoceros beetle which destroys coamnut trees. At the time this specimen wa, collected none of the characteristic signs of the raveges of this beetle were noticed in the coconut palms of Male Atoll, So it is probable that it had only been recently introduced. If this pest has become established in the Maldives it could prove to be a major disaster to the economy of the islands, which is largely dependent on the coconut palm. It is recommended that local authorities endeavor to determine if the rhinoceros beetle has become established, and that if it has vigorous efforts be made to eliminate it, before the infesta- tion reaches major proportions. A campaign to destroy all dead palms and other decaying vegetable trash which might serve as a breeding place would be a good start. Editors WH ‘evel 1) a ENS wy Ae Wey oS %, ——— el ee ee, Va os 2s ‘* \? SEN = oO aS ae Ha: Ph H i il a %, Oo Tou Ss Ss co S > “ml ~ -' ae €' Ore a oe cS : WB S O — SS 7, ae —— cr ay ih Op e | a : 3 oy _ =